schema.org/Person is a terribly designed format, that I would not bring into new projects. Here are some of the reasons:
1. It lacks diversity (e.g. GenderType is incompatible with German laws, requiring "diverse" gender in official forms, which is definitely not "unisex" defined in the format).
2. It looks very much like a US-centric God Object. It is not clear why DUNS, ISICv4 or NAICS are there, but other identifiers like national ID or SSN are not. It would be better to have a single "identifiers" key-value map instead of them, that would be extensible.
3. Contacts would deserve a dedicated structure and key-value map (why single telephone field? why messenger IDs not there?).
Yeah, it is - not to the level people envisioned years ago where people made up their own tags and found a blend of XML and HTML, but it absolutely is in place for more meaningful HTML. We have tags that tell you the purpose of content - header, nav, section, article, footer, etc. Those tags are read by screen readers and other accessibility tools, and do bring semantic value to those readers.
There is some neat application for example many tickets/hotel reservation/restaurant do include them and this allow to make some email clients displays these tickets in a standard way.
KDE Itinerary and Apple Wallet make use of this. For the KDE Itinerary part, you can read more on this here: https://volkerkrause.eu/ (look at the KDE Itinerary posts)
The EU is heavily investing in semantic data for education and digital identity, leaning on ontologies, RDF, json-ld and verifiable credentials.
Banks, research institutes, universities are joined in large consortiums. The German hotel reservation example above is one of the projects that’s getting attention.
If you get job where you need to merge graph data from a bunch of different sources, you will likely run into the semantic web stack. The semantic web stack is basically a bunch of technologies for building graph databases and sharing graph data across domains. It isn't some huge network of interlinked data sources like originally envisioned, but it is definitely not dead. I think the concept of graph databases is still something people are only beginning to discover the uses for. Most people only know of the initial XML serialisation of RDF and OWL which rightfully made people skeptical. Nowadays, the most common serialisation format is TURTLE which is basically just a subset of SPARQL and it is very readable.
In fact, the underlying technologies are all quite mature (triplestore implementations and the RDF, SPARQL, OWL, etc. standards) and used heavily in certain domains, such as biology and language. The standards are designed for easy serialisation and interoperability which is the antithesis of the "modern" walled garden and/or commercial API approach of the big Internet companies, although they also use semantic web tech to a certain extent.
In Europe, you will run into RDF quite often in academia (and research in general). I'm working on the newest version of the Danish WordNet, for example, which is published as RDF and OWL (in the turtle format) and which runs in Apache Jena. Coincidentally, I am actually developing this new version in Clojure which stole the triplestore concept from the semantic web stack and made it a mainstream database type in the Clojure ecosystem (Datomic, Datascript, Datahike, Datalevin, Asami).
EDI is only semi-standardized; in my experience usually each pair of trading partners requires a separate effort (usually the bigger partner imposes the requirement and the smaller one has to make changes). Some difficulties include:
* There are multiple protocols for exchanging EDI documents (AS1, SMTP, FTPS, SFTP...)
* There are multiple standards for the documents themselves (EDIFACT, X12, GS1 XML...)
* There are partner-specific business rules that need to be set up (this partner needs an ASN, that partner requires invoices to be for only one PO each, the other partner can only accept invoices which use their ERP’s internal codes...)
Some of these problems can be papered over with a “VAN” that can translate between standards, but I have yet to see one business send a non-PDF invoice to another without a lot of fuss.
Italy actually has a standard for electronic invoicing, and most companies are required to use it ("Fattura Elettronica"). It works by having a centralized portal to which invoices are submitted.
Why should there be? Resumes have to be read by humans to judge if they match what is needed. Checking boxes and matching buzzwords only gets you so far.
On the other hand, when you get a bit experienced, you have to filter your resume instead of just submitting everything, and the filters are different for every company, showing the things that are most relevant. E.g. some internship or award or early ten years ago should be included if the topic is relevant for the role and left out otherwise; the six different positions you had during your tenure at a large company may be important in one application to illustrate that you have done a particular role as well, and a waste of space for another application.
On the other hand, it makes all sense to have a standard "full" CV out of which to select the things to include.
We are weeding out based on superficial and incomplete data all the time. When you go to a supermarket with plenty of options to buy tomato sauce, you don't read the ingredient list of everyone of them. Maybe you do a first skim towards some "bio" tag section, or to the cheapest one. Once there you scan for known brands, and only when you get only two or three remaining options you read the fine print.
Same when choosing a partner: the first thing you use to filter are looks, assuming the risk of filter out a below average looking person that will be your perfect match.
We, of course, delude ourselves into thinking that our method is 100% objective and unbiased. But it never is. Even if you read each one of the 1,000 CVs, you won't read the first one with the same energy you read the 950, and it's likely that you anchor on a given CV that you read among the first 20, and everyone after that is meassured against it.
CVs are by themselves only a proxy to a real person: why don't skip the CV and just interview everyone? Nobody's life fits in two pages.
Makes it easier to spam them out when it's job-hunting time. On the HR side, they already filter on asinine keywords in the ATS. Why not join their game?
Note that there is some number of people who feel pretty strongly about obfuscating their age. I'm not sure what to feel about it personally but it's understandable.
I started getting better responses when reduced my work experience by 7 years ) It also let me remove some obsolete junk that interviewers never heard of ))
As one of the other replies has stated that trimming the older stuff from your resume can help. I took a variation on that by going into detail with jobs for the last 10 years or so and having a simple list of other places I've worked and what the job titles were. There's also a linkedin url which has the expanded version if the reader is interested in going back more but it let me keep my resume to a single page, single sided, summery of the "professional me".
I would have found it much more interesting and useful if the author had tried to sincerely answer or even ask the question of the title. Instead it’s just a complaint.
I guess the answer must be: Companies have no incentive to make it easy to apply to their job without manual intervention. Which makes sense -- if it were easy enough to apply to companies automatically, I guess some people would just write a script to apply to every company in their field, right?
The underlying problem is you need a JSON format about the company. But the things candidates care for generally are easy to lie about (good culture, career progression, even remote working options), so scatter guns are required. As they say recruitment is broken - but its not easily fixable because it’s broken because of game theory and human behaviour.
This is something of an issue with conference proposals as well. In fact, I've seen some conferences recently that put a strict limit on number of submittals. As a sometimes conference reviewer, I hate people throwing a bunch of overlapping and often generic stuff at the wall.
IMHO the author does answer the question without even noticing it.
At the end of the post, "The larger employers and job-hosting web sites need to get their acts together and work this out".
In a comment right below, the author says that this would be "giving job-seekers more control over their data".
Obviously, the larger employees and job-hosting web sites have no interest in allowing job-seekers more control over their data, quite the opposite, so they won't. And their active cooperation is absolutely required for such a data format to be adopted - so it won't be, at least not in a way that satisfies the author's wish of gaining more control over their data. That's it, it's that simple.
Parsing CVs is awful. We outsource this entire feature to Daxtra, who do a very average job (but, to be fair, better than anything we have time to write!).
There are two separate things going on in the linked post. One is the absence of a standard format as identified. The second is that HR software generally is enterprise software sold to leadership where usability is an afterthought, and so it only gets minimal development focus on having a quality resume ingestion algorithm. So you get the terrible parsing the author complains about. I'm confident that if any buyer really prioritized this feature being better, it would be.
I used a JSON resume once and I like the idea. It didn't help when I had to apply for an unemployment benefit (while still trying to enter the industry) and they asked for my resume as a word document.
As you can imagine, I felt like a clown trying to explain that I didn't have a word document because my resume was generated from a JSON file.
I did, however, have a PDF on a USB drive I always kept on me but they refused to accept USBs out of fear that I was trying to give them a virus. Eventually the lady processing my application gave up and printed the PDF off that was hosted on my website but also scolded me for not having a word doc.
The whole ordeal was pointless anyway since they said they can give me 70% of my rent.
tl;dr - HR is underfunded / under-resourced / un-aspirational.
Even if there were a turn-key software platform they all used, which natively supported a standard data format, they'd still find a way to screw it up.
I have my resume in markdown so I can create a word/pdf/html etc. with pandoc. It would be possible to write something that parses a json resume into markdown so it could be trivially format shifted
Json Resume is still going strong. I am one of the founders, I try to do a couple major maintenance periods per year. Currently I am working on updating all the community projects built, still got quite a few to add but currently there is -> https://jsonresume.org/projects/
I think over 3k+ people use the new Gist hosting. (In our old hosting we had around 10k resumes. Not including those who by pass the free community hosting)
===
On a personal note, I've loved having my resume in a standard;
- Depending on what type of company/person I am applying to I will change my theme on the fly. (Startup vibes I will make it look hipster, if it's a more formal role I will use a simple black and white theme)
- I use to lose my most recent resume constantly, having it in a Gist called resume.json that I just edit seems to have solved that for me.
- Hopefully one day a standard will get integration adoption so I can just upload my resume.json and not have to fill out the same form fields a hundred times.
Hey thanks for maintaining this project! I have been using it this round of applications and it’s just so much more convenient than trying to edit a historical doc(x) file with a rotating cast of word editors over the years.
Fair. I guess the reason I asked is because YAML is a little easier to play with for humans, especially since some sections need things like new lines among other things. Doing a YAML text block is easier to read and write than JSON's quoted string with new lines replaced with `\n`.
That's not really a "universal format", though. That's an "English language format". Actually, it's an "American format" since I see it specifies "organization" instead of "organisation".
Why should resumes in Vietnam or China be using English, after all.
In HTTP we have Get, Put, Post, Delete, Patch or Options and not obtenir, mettre, poster, supprimer, corriger, options. Yet the French use the web all the time, as do Chinese or Maui.
It would not be about how the fields are named, but about localised concepts. Maybe there are societies that have some educational concept which western Americans cannot fathom, but which is crucial on a CV there.
This,implies the standard needs to be flexible, extensible and localised. Which kindof defeats having a standard.
That's fair point. But normal French people aren't interacting with HTTP directly in any meaningful way. (More snarkily: they're using browsers built by Americans to connect to webservers built by Americans.) But I imagine that 99% of people who use json-resume today are interacting directly with it.
So I suppose that parallel suggests it depends on whether there's a realistic expectation of WYSIWYG editors for any kind of universal resume format. Which I think seems incredibly unlikely.
That's true of the fields in the JSON, but for what it's worth you can output your resume in templates for any language you like. I've updated mine to use Hiberno English for instance.
This is great, thank you for sharing. I love the idea of defining the data once and then having different templates and layouts to choose from. Very CSS Zen Garden.
Clearly written by an applicant rather than employer. Your resume is your chance to shine, use the opportunity or pay a service to do it for you.
The reason the author is trapped long enough in meat grinder hiring to notice this "problem" is likely precisely because of some indistinguishable cookie cutter bullet point soup getting them nowhere. If you can capture it in a data structure, it's not a resume!
See also: why isn't there a universal UI for web sites?
I've been hearing people react to current (American) college admissions by saying the problem is the Common Application allowing people to apply to many schools at once.
I would be more likely to locate the problem in increasing numbers of applicants rather than increasing numbers of applications per applicant, but I wouldn't be opposed to seeing a more centralized system go up. American medical schools accept students through such a centralized system. Chinese university admissions isn't centralized to the same degree as American medical schools, but it's still quite centralized, and more formalized.
Hiring companies might prefer to just have this as a "hoop."
There are, inevitably, some people who apply to a lot of jobs. Jobs they aren't qualified for. Low intent applications, where the applicant isn't really that interested. Etc.
Even if these are a minority, they apply to a lot of jobs. In any case, standardizing job applications (OP seems to be talking about application forms that are mostly resume-ish fields) just means more of these. More volume, more noise, probably not many more successful hires.
There's kind of the same dynamic on the other side. Most workers don't love the idea of submitting an indexable resume for employers to leaf through.
I agree that a hoop is needed - I recently posted a job on LinkedIn which has some kind of "click to apply" functionality, and it was clear that most of the applicants were just lazy clickers that probably hadn't even read the posting. On the other hand, if you make people jump through pointless procedural hoops, you're screening for people who are ok doing that.
Ask for a cover letter, you get people who actually want the job and hear from them why they are interested.
A cover letter is a kind of writing contest, or maybe it's perceived that way by a candidate. Writing about yourself can be stressful.
I'm not saying not to, ask for a cover letter. Just saying that this stuff is nuanced. Standardisation is great when you want very low friction. Job applications aren't frictionless, probably can't/shouldn't be. Some copy paste fields might be inelegant, but it's not changing the friction equation by much.
What's the point of a machine-compatible CV and a human-written cover letter? A human will need to sift through the same amount of information, except now they also need to read the long suck-up prose that you asked for.
The honest truth about why they want the job for most people is that they provide what you need and that they enjoy things like eating, having a roof over their head, and fun things to do with what's left over of the money you pay them.
Coming up with some drivel about how they've been entranced with Version_Five Inc's tech since they were twelve and how it's been their life-long dream to work with their famous Foobar Widget control language is a nice creative writing exercise but it hardly proves anything. If anything, it rewards lying and deceit to get hired quicker.
Many very capable people suck at writing, even if they are truly driven and interested, especially in the computer science field. A cover letter is a great screen for a job that involves a lot of writing, but for most jobs writing is only a side activity that shouldn't be treated as the main objective.
The thing is it's far more than a "modicum" in practice. There was a recent Harvard Business School study out there revealing that these systems are poorly calibrated (and/or these companies don't know how to use them), and as a result millions of candidates are getting rejected for reasons that are either specious (6-month resume "gaps" that may be due to normal life circumstances) or utterly meaningless reasons (like failing to match certain word combinations).
But I agree though that many HR departments would still love to use any tool that narrows the pool (however lossy the filter may be). Whether that is a good thing or bad thing or not is a matter of perspective.
In my view, it puts HR's own value and utility into question.
I very much like to write my resume the way I think represents me the best. The last think I need is to tailor it so some programmer who struggles with parsing.
When hire I also like to receive custom resume that helps me to understand better why hire. Again I am totally uninterested tailoring my offering to formalized set of checkboxes concocted by some "industry expert".
Documents like these shouldn't be automatically processed, they should be reviewed by humans. Reducing someone's life history to a list of educational institutions and employers feels robotic even for a software developer's mindset.
I understand that there are real life problems because companies do use automated processing on applications, but that kind of behaviour shouldn't be encouraged.
Downside is when you have to make slight changes each time to better emphasize some aspect of your experience, depending on the job applied for. Recruiting agencies often do this to sex up their candidates chance of winning the job lottery.
I wonder if there's any advantage in using fonts and layout that are adversarial to automated resume processing, as in if something fails to scan then perhaps a human is more likely to actually look at it. But perhaps HR just throws those out.
Does anyone OCR? The systems I've heard of just extract the text from PDFs/docs. Then if some bits cannot be extracted, I was asked to type them myself.
Quaint to think so, but no. If a system is using automated scanning and the document doesn’t scan correctly the recruiter is not going through the ones that didn’t work.
I’d wager in many cases they aren’t even aware of an failure to process a resume occurring.
Also in the eventual human review if you have made yours illegible on purpose, you probably lose any advantage you gained with passing te very low bar of automatic filtering
That's already the case. You're much more likely to get hired if you know someone on the inside, if you've been recommended. If you apply through the official channels, you might not even get a response.
But I think an official format for CVs might not work so well because CVs in different fields aren't formatted in the same way. A programmer isn't going to put forward the same kinds of things as a musician or a professor in psychology.
As a hiring manager you’d be inclined to disagree.
If resumes could be standardized and easily searchable for information it would help find candidates much more accurately and quickly without having to read.
Even better would be if there was a verification standard where you could get a blue checkmark on your resume meaning all the information is accurate and not made up bullshit, then you could limit searches only to verified resumes and do less investigation or third party background checking.
> If resumes could be standardized and easily searchable for information it would help find candidates much more accurately and quickly without having to read.
I see how that can be useful from the recruiting side of the table, but fail to see how the candidate benefits.
Think back to the first time you wrote a resume. For me and all of my college buddies at least, it was super stressful, because we had no idea what it was supposed to look like and nothing but vague rumors to go on. If there were a straightforward canonical answer for how my resume should be formatted it would have saved me quite a bit of work and quite a bit of stress that I was doing it wrong.
A standard requires people to agree. When I was in college, I was told that a resume was a list of buzzwords and past jobs. 1 page max. That got me quite a few jobs. When I interviewed at Google, people were like "this is it?" when looking at my resume (not that it mattered). I started doing interviews and saw the resumes that ended up at Google and they were all multi-page affairs with details about all the projects you worked on at a job. I started doing that, only to be told "wow, that's a really long resume". (Hint: if your role requires 10 years of experience, applicants are probably going to have a long resume!)
The key point here, I guess, is that a lot of people have told me my resume is in the wrong format, but I still got the job. What's the right format for one employer is the wrong format for another employer. So there is no way to win. I now have no clue how to write a resume, but it also doesn't seem to matter.
(Still bitter about how few people ever look at my Github, though. People put that as a custom field in their automated system, and then the interviewers have nothing to say about my projects!)
That sounds like it pushes hiring managers forward, not industries. You get a benefit out of making it easier to fill open spaces, at the cost of everyone else's unique life experiences.
We have a "verification standard" in the form of diplomas and certificates and neither of those have fixed the hiring issue so far.
If you want to skip background checking, just make people bring their proof of certification with them to the job interview. Of course, those certifications barely mean anything in most fields of work, but it's an exact equivalent of the blue checkmark system you propose.
I get that hiring is hard, but that's why hiring managers exist in the first place. If we used a nicely standardised, automatically validated system, all recruiters and hiring managers would be out of a job. Why pay someone to do that stuff when you could pay a cheap machine learned AI to fill a list of requirements for you?
You could probably only try to verify local education - for the foreign education not that easy. Even employment (even local) is very hard if possible at all. For example, I work for a large corp X as a contractor. It pays my agency Y. Y pays my umbrella company Z. Z pays me as its employee. There is hardly an easy way to link me to company X yet I've been there for 9 years.
I see what you're driving at -- but that implementing that "verification standard" (i.e. standardized, continually updating 3rd-party background checks) would be a huge undertaking, and a larger than cottage-sized industry in itself.
Way, way more than a matter of parsing a bunch of standardized resume documents.
>Documents like these shouldn't be automatically processed, they should be reviewed by humans.
That's very nice to think about but in the real world if people don't match some exact keywords they get thrown away. HR is looking for "node.js" but you wrote it like "nodejs"? Though luck pal, bye. HR wants a "computer scientist" but you have a degree in "computer science"? Same, automatically discarded.
If anything, a standard format would at least allow people to classify themselves/others correctly and without ambiguity.
So what? You're looking at it one job at a time, and for any one given job it might indeed suck. But collectively, the times when you're unfairly tossed out are balanced by the times when other people are, and you benefit from being in a smaller pool. If you spell it "nodejs," surely some HR person somewhere does too.
It's random, and not based on any job-relevant differences, for sure. But hiring is always going to be pretty random so long as we're pretty bad at predicting who'll be good at any given job. So long as there's no systemic bias, it's fine, and selection by idiosyncratic spelling is at least much less biased than a lot of other irrelevant criteria that get used in hiring.
> That's very nice to think about but in the real world if people don't match some exact keywords they get thrown away. HR is looking for "node.js" but you wrote it like "nodejs"? Though luck pal, bye. HR wants a "computer scientist" but you have a degree in "computer science"? Same, automatically discarded
I'm sure this happened somewhere at some point in time, but IMO it's becoming a myth blown out of proportion.
I've never actually worked at a tech company that had any sort of automatic resume filtering software. As a hiring manager, we didn't even have HR pre-screening resumes for us. If we did have recruiters, they were competent and incentivized to find good candidates (e.g. wouldn't be unnecessarily rejecting good candidates because they were missing exact spellings of keywords)
It would seem ludicrous in 2022 and this hiring crunch for a tech company (one that you'd actually want to work for) to arbitrarily filter out resumes based on incomplete keyword searching.
I co-founded an ATS company and I can say that we were never asked for, nor provided, any functionality to automatically screen candidates out based on keyword filters (though candidate answers to actual questions like "Do you 5 years experience with node.js?" are often used to automatically screen out).
That said, I have heard this often enough that some people must do it. I can't imagine why though, as it takes only a few seconds for a human to scan a resume. Even at 10 seconds per resume (a conservative estimate), that means the human could screen 360 resumes per hour.
Where do these resumes in the ATS come from? Sourcing tends to use searches or keywords, which “filter” by not selecting in the first place. Applicant trackers tend to be a few steps downstream of that, so wouldn’t need to filter.
Based on experience at firms with on the order of 1:1 applicants to hires up to 10,000:1 applicants to hires, I’d agree most such tools are not present at most companies, as they tend to be used from ratios of 100:1 and up.
On the other hand, while most companies might not have or use such tools, the more jobs a company has, the more likely they are to have some form of selection or filtering in the pipeline.
I’ve interviewed the CEOs of many of the largest ATSs for Ladders News, and while there is no automated system, the tools do exist for a company to screen in this basis.
But none of the ATSs really collect great data on how many of their customers use them in their core processes. From observing actual, heavy duty recruiter and HR usage at Ladders, I suspect it is not very many.
These are people people after all, and they are not great at writing REGEX style queries to find only the precisely correct resumes they are looking for. You also need to remember that they are frequently working on novel searches each week, and wouldn’t have the confidence or the capability to craft such searches.
I'm not sure if I'd mind that. If HR is made up of robots that stupid, that sounds like a terrible company to work for.
A standard format wouldn't solve the problems you described. There's no comprehensive list of frameworks and programming languages, let alone for types of experience outside computer science. The data would be structured, but HR would still throw out the bits of data that need to be filled in manually.
It varies. There is a reason there is some level dislike for HR cohorts. Keyword bingo is one of those things.
Minor anecdote incoming. I applied to a company A. It was a fit. I think I used the right keywords. Zero response. I talked with my buddy there. She talked with hiring manager. I got hired by company A. I later found out HR was unhappy, because it undermined their 'process' - such as it is.
There just has to be a better way to do this when compared to things as they are now.
Do you really care that you made HR unhappy though? Seems like you shouldn't, the company/team management is probably happy to pursue references over the funnel of random candidates, and the unhappy HR staffer is better off learning from the experience.
Not really is the truthful answer. The job in question had my best boss so far, so upsetting HR drone was a necessary evil in retrospect. I just dislike making people upset if it can be avoided. In my line of work, I do not get to make a ton of friends along the way. And who knows where that drone might end up in 10 years.
Do you really care that you made HR unhappy though?
Fundamentally it does kind of suck to have to put up people in your environment who are "unhappy" with you, or the fact that you snuck in there -- not because you don't belong there, but because of their own basic ignorance as to how things actually work.
Not that you can't grit your teeth and live with it - it's a job after all, and that's what you get paid for. But still, it does kind of suck.
Going direct to a hiring manager is always the best route to the job.
Any standardized process will lack fidelity and excellence in specific cases. Or maybe even in all cases, when the company has made the trade off for cost or speed over quality.
But that’s what big company life is like, and that’s why they have standardized processes. Because even with the errors, in aggregate, it works out better for the business.
> HR is looking for "node.js" but you wrote it like "nodejs"? Though luck pal, bye. HR wants a "computer scientist" but you have a degree in "computer science"? Same, automatically discarded.
Wouldn't those issues be more likely to occur as a result of automated processing? I imagine that most humans who deal with technical hiring would guess that "node.js" and "nodejs" are probably the same thing.
I believe I understand your point – therefore I would like to take this opportunity to channel something that 'patio11 wrote on Twitter many years ago:
“Do not send in a job application before you have an actual human
being who has expressed enthusiasm for reading or forwarding it.”[0]
I believe this is sound advice, and the pitfalls you mentioned was probably the reasons that he gave this advice.
That's a good advice but probably for the top X percent in their industry who have a luxury to choose their employer - not everyone is in this position.
It's not about picking your employer so much as it is about getting past the outer layer and in touch with an interested human who can champion you through the rest of the process. In the absence of a pre-existing relationship, a good recruiter is invaluable for this.
Recruiters can definitely be a good way. That’s how I landed, like, nearly every job I have ever been offered.
Networking and reaching out to people at interesting companies is another way.
You will probably still have to write a cover letter and send in your résumé. But if you have someone on the inside who will mention to the hiring manager that, ”hey, you should check out Jane’s application” it will probably make a great deal of difference compared to just applying and hoping for the best.
And please folks, remember that FAANG is not the only game in town. There are lots and lots and lots of other companies out there, and many of them also pay well since they need to compete with FAANG to at least some extent.
I remember the HR chief of [famous company] complaining to me that she'd already been ordered to offer me a job, and needed a resume to check her bureaucratic boxen.
I have to agree, I have a lot of respect for McKenzie but this advice seems unwise. I've landed four jobs (including my longest-lasting gig to date) by sending in resumes to job listings without knowing anyone internally.
I’m curious if you have much evidence that this is common? I suspect it’s mostly a myth that is easy to believe (it can feel good to believe it because it means you were removed from the applicant pipeline because of stupid computer reasons rather than because you don’t appear to meet some bar to a human, even if that bar is a stupid thing like prestige) and oft repeated.
I definitely think it’s likely that most people on hn don’t work for companies that do this, but maybe hn just skews toward certain kinds of tech jobs.
In college I got the advice that if you couldn’t highlight text on your resume then a bot can’t read it and it might get thrown out. And if that might cause it why risk it?
I think the fact it’s low cost to avoid and high cost if it occurs keeps most people just doing it regardless of whether it’s true or not
It happens in big companies. My team is hiring about 100 people in the next 6 months.
For external candidates, we see less than 30% of all resumes. As with all things, 80% of the rest are junk and another 20% are potentially viable candidates.
It would seem most of the over-confident “it’s a myth” comments in this thread are from folks not familiar with typical processes for hiring in the 100s to 1000s.
You shouldn't be hiring that many. The company might, but the managers and people doing the interviews should be looking at maybe 20 positions per year. They can read résumés themselves and give it some thought.
Who sorts which 20 out of 20,000 resumes go to which managers? Not the managers. Who is “the company”? That’s who is keyword searching/matching/filtering.
The hiring manager. We have got tired of hr rejecting great résumés because they don't understand keywords. They have no clue that qt and qt5 are the same thing, and if we put both keywords on they reject someone with only one. Nor do they get that someone great with c++ is almost automatically great at qt. I'd be happy to teach c++ to someone with rust or Haskell experience if they are willing to learn. (And if it is rust we are open to new work in rust, though we have a lot of legacy c++ you will maintain)
We consider ourselves lucky to get 50 résumés, and if even one isn't obviously disqualified that is a miracle
That's very nice to think about but in the real world if people don't match some exact keywords they get thrown away.
This isn't the "real world" you're referring to here -- it's the fin-de-big-tech bubble world where everyone and their dog is not only sold, smack-down drunk on the idea that Algorithms and Data are the Solution to Your Problems. They aren't of course -- it's just a giant hornswoggle. This "world" needs to end, and it needs to end soon.
My belief is that it will. But until it does, companies that run crappy ATSs (like there's any other kind) deserve the buzzword-gurgling, keyword-dropping candidate pool get. And developers who get "rejected" by these companies should be grateful for the sublime gift of this rejection -- and for the opportunity to laser focus on companies and teams that use their heads to hire, rather than a fleet of bots.
Snark aside - hiring managers that actually read resumes (and yes, actually at least skim each and every one -- really it ain't that hard) are golden to work for (other factors being equal). Really, you don't want to waste your time with companies that have drunken the ATS kool-aid. Really you don't.
That said, however:
HR is looking for "node.js" but you wrote it like "nodejs"? Though luck pal, bye.
It's "Node.js" with a dot. If you spell it "nodejs", that's a serious red flag and you shouldn't be surprised if it gets you automatically discarded -- even by a human reviewer.
That's what resumes used to be, after all -- a kind of a take-home test where you really do have enough time (and perfect knowledge) to get everything 100 percent right. And also a test of your awareness of the fact that, yes, in critical business communications at least -- this level of correctness does matter, and it matters a lot.
Google has 411m hits for Node.js and 66m for nodejs. I'd guess somewhere in the 66 million people calling it nodejs, we could find a real diamond in the rough if we heard them out at the next interview!
That's exactly what I mean -- people believe in Big Data so much these days that when something appears in a sufficient portion of search results, it must have actual validity.
Isn't that the same as natural language? "Bone Apple Tea" gets corrected both because it's wrong, and because many people know so. "Begging the Question" is used incorrectly _so frequently_ that even dictionaries have come to accept the incorrect usage as an alternative. It's now correct-ish just because it's popular.
So if Node.js has that many more hits than nodejs, it's not surprising that some people see it as canonical.
Hearing the context of this conversation, though, I can see that "nodejs" in your resume is a signal to senior peers within that community--others who are "in the know". That may be a useful filter.
As an aside, I never thought "begging the question" is used incorrectly because it is easy to tell whether or not someone is referring to the fallacy based on the context. If the meaning can be understood, then that's what matters.
> That's what resumes used to be, after all -- a kind of a take-home test where you really do have enough time (and perfect knowledge) to get everything 100 percent right.
The only thing this filters for is people who have been looking for a job for a long time and/or are people focused on presentation instead of content, good luck with that.
We’ve hired incredible engineers across that have had typos and major resume mistakes many times and they had no communications problems on the job.
A resume is a background/interests check for the hiring manager. Giving them a spell check is about as useful as measuring the ratio of times the letter ‘a’ appears relative to ‘e’.
> critical business communications at least -- this level of correctness does matter, and it matters a lot.
This sounds like someone who is unable to contextualize when that would be relevant. Unless you’re writing user documentation there are zero times in an engineer’s day-to-day communications where writing nodejs instead of “Node.js” would have any impact at all.
Giving them a spell check is about as useful as measuring the ratio of times the letter ‘a’ appears relative to ‘e’.
I'll leave the matter of whether one should be expect resumes to be at least reasonably correct (in terms of spelling and grammar) or not - or what the threshold for "reasonably" should be on this front - to the side for now. It's obviously a question of taste, to a certain degree.
But sorry - what you're saying about the ratio of 'a' to 'e' makes no sense at all.
And on this front:
Unless you’re writing user documentation
I don't know what planet you live on -- but on my planet, good engineers write "user documentation" (in the sense of READMEs, docstrings, etc) and participate in mailing lists, do conference presentations, etc, all the time. And yes, while nobody is 100 percent perfect, consistently making a large number of gaffes (or more to the point: conveying an attitude that these things just don't matter; or that you're so hot that they shouldn't matter for you) in these contexts -- particularly when there there is plenty of time available to edit and review -- is definitely seen as a red flag, both in and outside the company.
These days I think I work at the best team I ever worked on and it seems I'm in the top two in spelling, it is almost as if the rest if the team is mildly dyslexic or doesn't care.
But they are nice to work with, smart and they make good code.
I'll take them 49 times out of 50 over some perfect in code and writing but nasty in person ex colleagues.
Even if I really really really need a specialist on hight performance <whatever it was, I can't add a dead giveaway here> I'd still try to look elsewhere first.
Edit:
... and those two ex colleagues? Their code was good but it wasn't perfect either.
Am 100 percent with you there - nice and basically decent to work with always trumps the willful disregard for these basic qualities. No matter how skilled they are, or their accomplishments.
> But sorry - what you're saying about the ratio of 'a' to 'e' makes no sense at all.
Sorry I confused you. Let me be clear, it’s not relevant at all to job performance. I would take someone with terrible spelling and punctuation who actually writes documentation over someone who writes half as much because they are obsessing over details like that.
> in the sense of READMEs, docstrings, et
These are generally for other developers. Developers do not get confused when they see “NodeJs” instead of “Node.js”. Do you know what point you were trying to make?
> is definitely seen as a red flag, both in and outside the company.
> For any company you'd want to work for, that is.
Not sure what kind of company you’re talking about, but it’s definitely not a tech company. Maybe Wall Street or something a little slower paced?
It's not about communication, or spelling, or grammar, it's about attention to detail.
I've seen my fair share of resumes and hired my fair share of people, and I've come to the conclusion that sloppy, error-ridden, CVs are a sign of a lack of attention to detail.
Here's a candidate with all the time in the world, and huge motivation, and they still can't complete this task correctly. It's not a good look.
I'm not saying a single error I'd a deal-breaker, but it puts the applicant on the back foot. No doubt many of my rejects are quality engineers, but engineering, and software development are all about details.
A job hunt typically takes weeks or months of calendar time, and dozens of hours of clock time. So if a candidate has good management of his time and effort, he can budget one of those hours to improve the attention to detail of his CV.
The motivation might not be "huge" at all, considering you might not be expecting any response, might have done 10 of these in the last hour, pay may not even be acceptable, etc.
> Here's a candidate with all the time in the world, and huge motivation,
You are woefully out of touch with the job market. And I’m saying this as someone who worked as a hiring manager at one of the most applied to tech companies in the world.
Its not just big tech that does this. HR is also incompetent in academia and has similar practices at many institutions
Edit: oh, and also US state governments. there's a reason they're known for being incompetent, people job hop within months and get keywords on their resumes and nothing ever gets done
>It's "Node.js" with a dot. If you spell it "nodejs", that's a serious red flag and you shouldn't be surprised if it gets you automatically discarded -- even by a human reviewer.
Oh, really!? I'll be very glad NOT to work at a company that throws aways candidates because of such triviliaties.
Perfectionism and pedantism are good qualities some teams are after, it is unsurprising that some people would reject a candidate who did not bother “properly” writing the name of the technology they use.
Something tells me they could easily focus their attention on something not necessarily related to the most critical parts of the project. They could be very pedantic (and often are) about some irrelevant thing.
nodejs came out on 2009, I've used it non-stop since then and probably built close to 100 things with it. I am on the author's list of both node and v8.
I write it as nodejs because that's how I got used to it on the mailing lists and several places that I frequent.
If a "team" decides not to hire me because of the latter, then ... thanks, I guess?
I guess I'll backpedal at this point and acknowledge that that "nodejs" may in fact be not only a correct and accepted spelling -- but in fact a hip "in-the-know" spelling -- within a certain community.
But at this point we're getting into the territory of what's seen as a positive in some niche communities may not be seen as a positive signal outside that community. And in business communications, what masters most is knowing who your audience is, and how your message is received.
Then you being outside the community, have done nothing but filter out the best candidates by focusing on the wrong details under the banner of “attention to detail”
It’s ego.
What is the résumé’s job? To convey experience and work in a positive light. A small typo that may incidentally be the correct spelling shouldn’t derail that if the goal is to hire the most talented people.
To my defense, I was going by my Bayesian prior for certain language names for which one could be 100 percent sure there was no alternate accepted spelling or capitalization.
So if one put JAVA or PERL on their resume, that would generate a definite "pass" signal.
Agree or disagree, this is great feedback for job seekers. Spelling is often used as a filter.
What are some other signals that act as red flags for a potential applicant in your hiring flow?
Well another major turn-off would be: a "primary skills" section that mentions 30 different random applications / protocols that one might have been incidentally exposed to for a week or 2 at some job ... 15 years ago. No one can be expert-level in all of these things. Just tell us the top 3-6 that really matter and which define you.
Or even if it's just 1 or 2. It's infinitely better to hire someone who is in fact really solid at what they say they know, than someone who tries to spam you with every random keyword they've been exposed to in the hope that you won't drill down and ask them any hard questions about most of them.
That, and skill listings that don't pass the "apples and orange" test. For example:
I have a lot of merge requests in some pretty big projects correcting SASS to Sass in the code and documentation. I'm surprised no one ever seems to question their notion of spelling proper names when it's a quick web search away.
Even back in the era when reading resumes and hiring was a manual human process, the standard advice was that resumes should be perfect. Nicely formatted, with no spelling or grammatical errors. You have ample time to think about the content, to proofread it yourself, and to have someone else proofread it. It's the first impression the potential employer has of you, no reason to not make an effort.
I'm not sure that's true at all. Coming out of school pre-Web when real information was much harder to come by (even job postings), I sprayed and prayed a lot of resumes to HR departments. (And my first engineering job came through one of those.)
> It's "Node.js" with a dot. If you spell it "nodejs", that's a serious red flag and you shouldn't be surprised if it gets you automatically discarded -- even by a human reviewer.
Yikes. “It’s not JavaScript it’s Ecmascript! What a RED FLAG.”
I’d say they are. Who cares if I write “Nodejs” or “Node.js”? The reader knows exactly what I meant, and that’s what matters. Anything else is just needless pedantry.
It really does require an eye for detail (in places where it matters) to a degree that comes epsilon close to being "pedantic" in the annoying sense -- without quite going over that line.
It's not about communication, here. You want to be absolutely perfect in your CV, to prove that you've spent time and effort making it perfect. It serves no other purpose but to show your determination. If communication was the point, a long rant in which you write "yeah I worked with node.js, it was cool lol" would be OK as well, after all it gets the point across doesn't it?
That's fine. I see it like this: If there is a stupid mistake, it shows that I am a fallible human after all. (I tend to think that I don't have stupid mistakes in my CV)
But my CV also shows part of my character, mainly the fun part. If that is deemed unprofessional by a company, I immediately know they are no good fit for me.
The thing is: As with every relationship, they are two way streets. I am not a beggar as a potential employee, neither am I king. We see eye to eye or not at all.
The things is -- they ding you for a whole range of other things, all the time. Like not working for the right kind of company in your last job, failing their bullshit tests, or heaven forbid, having a "gap" in your resume. As if having a life (or a family member that needs taking care of) is to be seen as thumbing your nose at their sacred cause.
At least fixing stupid mistakes in your resume is well within your control, and (unlike cramming for their tests) doesn't take hours and hours of your time.
I'm an OCD-ish perfectionist (especially in writing) if the world has ever seen one, to the point where I didn't stark working on my bachelor's or my master's thesis before I had exactly the markdown + latex setup + custom typography that I wanted. So I would definitely internally cringe if I'd see typos like this.
But to reject a candidate just because they're not the same kind of pedant that I am would be quite unreasonable IMHO. As long as the whole thing isn't littered with typos or formatting issues, that's just something I wouldn't pay too much attention to. Not everyone needs to be the person to dot every i and cross every t.
Also, I really want to see the HR department that can confidently reject or even mark down a candidate when most of the writing I've seen from most people in any company I've ever been (including from HR departments, PMs or executives) has been "meh" at best, and full of errors or at least awkward language more often than not. For better or worse, writing skills are not necessarily something we tend to select for in the industry.
The difference is that JavaScript is the name of an actual language. Where nodejs isn't the actual name of the project. The Node.js project advertises itself as a "JavaScript runtime." Both JavaScript and EMCAScript are acceptable.
Absolutely. We typically receive hundreds of resumes for every position. The easiest way to winnow them down is to throw out any with spelling or grammar errors. It is astonishing how high a percentage include these errors. There is absolutely no excuse to have spelling and grammar errors in your resume if you are serious about trying to find work.
Speaking of red squiggly lines in Word/MS products. Is it just me or has the grammar check gotten incredibly worse the past 2 years? It keeps complaining about odd grammar forms and suggesting I replace perfectly valid word-structures with shorter ones? The one I encounter most frequently is the suggestion to replace "in order to" with just "to", which is incredibly frustrating as they mean two different things entirely.
All of the lines below have red squiggly lines under them in my browser's text area:
- Qt Quick
- Datalog
- Datomic
- TensorFlow
> I've noticed that ESL speakers make different kinds of mistakes than sloppy native speakers do.
I've never seen a non-native speaker writing "would of" or using "they're" instead of "their". On the other hand I can't count how many times I've skipped an article or used a wrong one somewhere (since my native language doesn't use them at all).
Yup. My wife learned her first word of English at 43, I've had plenty of time to see how one's mother tongue has a permanent effect on language. My interpretation of what I see:
She has internal word concepts. When the English words fit cleanly into her existing concepts it's easy. When two English words map into one word concept it's considerably harder, it took her a long time to correctly separate turn on from open (and turn off from close), but at least any given thing always goes the same way. In situations where that doesn't even work she will make frequent errors even to this day--do not trust her use of gender words.
It’s a document sent to tons of people. Revised by tons of hiring managers. Can’t you spend 5 minute looking at a dictionary? If you cant be bothered to spend 5 minutes researching this for an important document, what makes me think that you will spend 5 minutes trying to be good at your job. Lack of care on your resume is lack of care on lots of other things.
Companies don't seem to treat it like an important document. I can't tell you how many times an interviewer seemed to be reading my resume for the very first time during the interview itself...
As the interviewer, in several places I've worked, I'm usually handed the resume 10 minutes before the interview along with a couple of questions my manager wants me to ask. By that point, it's gone through HR, managers, etc. You're getting hired if you can get past me and whomever I'm interviewing with.
There was even one time they forgot to give me the resume... that was probably the funniest interview ever. They got the job too.
On the flip side, I do technical interviews (I am new to it though), and I spend about half an hour preparing for an interview including going through the CV that made it to me (and making notes/preparing questions) and researching the candidate based on what I see in it (e.g following links, etc)
I do the same thing. The CV is typically made available to me a couple of days before the interview, and I usually take a look at it the day before. Obviously I can't really know how carefully it gets scanned before they decide to arrange the technical interview (which is where I step in).
If you cant be bothered to spend 5 minutes researching this for an important document, what makes me think that you will spend 5 minutes trying to be good at your job
Bullshit. There are many things person may care about and many things they may not. If this position is unrelated to professional writing, it's just irrelevant. Are you questioning their abstract ability to "care" and trying to translate spelling care to job care? Bad news, it doesn't translate (aside from the fact that ideal perfectionists can't get shit done ever; your best candidate still figures out the ideal form of his resume and is too anxious to click "publish").
Some people aren't native speakers. Do they really need to know how to spell perfectly to work for you?
Perfectly, no. But in formal communications, they need to show that they're at least making an effort to get things reasonably correct. And if it matters enough, you can always find a native-speaking friend to review your resume or cover letter for you.
(Source: been there, done that, applying for jobs in other countries).
We rarely receive resumes from people who are not native speakers. It is the native speaker who make the worst spelling and grammar errors. It is so strange in this era when every computer has a spell checker and grammar checker. They aren’t perfect, but most of the errors are so egregious that they clearly aren’t even trying.
I think it's perfectly reasonable to want to filter out a candidate that can't be bothered to spent 3 minutes and run their resume through a spell checker. It tells you something important about the candidate.
Of course, it is also reasonable to be forgiving on resume errors. It depends on the culture and team you're trying to build.
Native speakers are the worst at pointing typos in my experience because they are so sure of themselves.
Literally every time I have submitted an (academic) paper at least one US reviewer had to point out I made several spelling mistakes when I actually used the British spelling (consistently) instead of the US spelling.
Sorry my man but it is called "English", I think I am allowed to spell things like they do in England.
These are true:
- hiring managers that actually read resumes are golden to work for
- a test of your awareness of the fact that, yes, in critical business communications at least -- this level of correctness does matter, and it matters a lot.
And for people who want to work in startup land, it’s an ideal.
Once a company gets past 40 or so people, though, ATS adoption increases. And 150+ employees without an ATS are rare. It’s just that the volume of candidates and interview processes that need to be managed require there be some software to manage the workflow.
So, when I see the ads with XML misspelled XLM like XLM and XSLT experience wanted - or similar stuff that happens every now and then - I should spin up a new CV and replace instances of XML with XLM?
on edit: before people come in with not wanting to work for a company that would make such a stupid mistake, first off I believe mistakes sometimes happen, and second Henry Thompson leader of the XML Schema working group, had the subtitle of his LinkedIn page saying something like he was an expert in XLM for at least a year (I should really have taken a screen shot) so I mean I think it would just as silly to not work someplace because they made a mistake as it would for them to screen you out for making a mistake.
Well said. It'll get to a human eventually, but the first step is pretty robotic anyway human or not. Let's dig up the legendary Facebook recruiter specifically looking for Unix experience.
The former I think you’re spot-on. If you’re serious about hiring, some qualified to do so needs to be reading resumes.
The latter is complicated. I tend to agree that the industry has over-pivoted to “dynamic programming whiteboard puzzle” as the LSAT of high-paying software work. But while most seem to agree that leetcode interviews suck in tons of ways, something that is unambiguously superior seems to be an open problem.
And some algorithms and data structure testing is important for hard software work. Even choosing the right library requires some basics. Is every SWE at Amazon implementing CS papers every month? No, clearly not. But unless machine resources are free in your domain, brute-force doesn’t always work.
FANG people do in fact get elitist about their CMU educations. But there’s another crowd that throws the baby out with the bath water: “I solve business problems, not reverse linked lists.” Well, what kind of business problems don’t involve some computation?
It’s a balance that our industry seems to still be trying to find.
I kinda agree with you -- half the job as a programmer is being really tedious and paying attention to written text on a screen. If you can't reasonably review your resume and use spell check, it is not a great look.
> It's "Node.js" with a dot. If you spell it "nodejs", that's a serious red flag and you shouldn't be surprised if it gets you automatically discarded -- even by a human reviewer.
Best example of someone I wouldn't want to work for ever
The question is does that really bring something good, either way, or is it just a whim. Some HR check the shape of your ears, and if it doesn't match some patterns, they skip.
perfect filter, both ways
Two ways of three. Some innocent person may accidentally put Node.js in their resume and get hired.
>> It's "Node.js" with a dot. If you spell it "nodejs", that's a serious red flag and you shouldn't be surprised if it gets you automatically discarded -- even by a human reviewer.
I review every single resume and applicant I hire. My experience tells me talent acquisition professionals don’t really know how to match and will often suggest poor matches and not bring forward odd fits with interesting backgrounds.
I get hundreds posting into some of the roles I’ve hired for and I’ve got to look at every single one in order to find a possible match. It generally takes me about 30 seconds to a minute to assess a resume. A kernel of interest stands out quickly or a smoke screen of buzz words tells me it’s the wrong candidate.
Algorithms aren’t a substitute for an active hiring manager being really interested in finding the right candidate to join their team.
This is why applying for jobs using a resume is a losing battle. Make connections, get introductions, get recommendations. Plenty of ways to explore new jobs without being resume 123 out of 500 who applied. No telling what arbitrary filtering criteria will knock you out of the competition. Absolutely no feedback on why you weren't considered.
> If anything, a standard format would at least allow people to classify themselves/others correctly and without ambiguity.
Except it doesn't, because a lot of things in life doesn't fit in that boxes that well.
You might not have experience in node.js, but in JS and some of the node.js-alike server side VMs.
Even just a "degree in computer science" can mean many things. Furthermore you might have something which for the position is "equivalent", like a "master of science in cognitive systems with a specialization on AI" which contained a lot of CS courses, and maybe also some pr-axis experience. Or you just started working as a programming free lancer with 16. Or ...
Furthermore a resume is quite individualistic, beyond just the raw content it contains. How that content is represented can sometimes give you hints, about what kind of questions you want to ask during an interview.
I have seen a bunch of standardized forms for resumes some companies opted into, they _always_ caused endless problems.
HR is looking for "node.js" but you wrote it like "nodejs"? Though luck pal, bye. HR wants a "computer scientist" but you have a degree in "computer science"? Same, automatically discarded.
Fixing that would require not only a universal data format but also a universal taxonomy of terms to describe skills. That would need everyone to agree on how to describe what they do. Easy enough when you're describing programming languages, but effectively impossible for any softer skills. This is precisely why you need the human aspect.
The way to fix the issue you describe is to educate HR people, not to try to apply technology to a non-tech problem.
I'm pretty sure my Resume says "crazy outside-the-box madman" on it somewhere near the top. It pretty much guarantees that I work somewhere with a sense of humor while rejected from places that take themselves too seriously.
Resumes aren't just a list of skills and work history (education isn't even on my resume anymore), it's also a chance to let a small amount of your personality show through so you stand out against the pile.
Of course, if you're submitting to a big corp, you'll be fighting against AI reading your resume first, so, write for that if that's how you roll. I usually only apply to smaller companies that apparently end up as bigger companies by the time I move on.
My LinkedIn recruiter spam dropped substantially when I added "literally a wizard" to the skills section (and removed everything remotely tech-telated).
> The way to fix the issue you describe is to educate HR people, not to try to apply technology to a non-tech problem.
I wish it was easier to do. At my first job after university (not my first job altogether) we initially had great rapport between "DCOps" and HR. To be honest, our group was rather... irreverent, which made it even more interesting that we had such great rapport (we had all the stereotypes going for us, including female-only HR team and basement-dwelling male-only sysadmins).
But thanks to great cooperation, HR would do basic filtering on things they actually checked with us about, then forwarded us the CVs unmodified to give our own opinion before deciding whether to go with it or not.
Unfortunately some time later head of HR ended up let go due to some conflict with C-level I believe, and things slowly reverted to the mean :(
Fixing that would require not only a universal data format but also a universal taxonomy of terms
A regular HR can only do CTRL-F CTRL-V efficiently in their PDF viewer, when automated system would likely test for /\bnode(?:\.?js)?\b/i, because it's a result of a development cycle, which usually includes some field analysis. HRs would LOVE to select something like "where has($node) and years($redux) >= 2 order by years($php) desc". They are not idiots, they just have piles of data and nonsensical tools. Of course they optimize for nonsense.
A human could search for different variations too, but they are average in average and have no the expertise to install a regex-able viewer. It's stupid little things like this that prevent them from doing their job better, not some systemic issue of their own.
I can see why. Of course a role that requires CTRL-F level of skills attracts low-skill-required resources to a "helper" subrole. But if you replace it with automation, true skills will remain and thrive. Not that I like them on emotional level, their job is to hire you as cheap as possible and demand as much as they can. But hey it's a market. Learn to negotiate and to plan ahead financially for search periods to avoid premature decision pitfalls.
Maybe I've not applied and worked with big enough companies, but I've never seen an HR person be helpful or in any way been aligned with my interests. This could be forgiven as they may well serve interests which are opposite to mine at times, but there are numerous times where HR representatives just (from what I can gather) pick suboptimal strategies to try and convince me to do something, be it remain or take up a job offer or anything else, i.e. I've yet to see an HR person successfully manage a human resource in an optimal way. And by that I mean that statistically the outcomes would be no different if it was a random person with normal communication skills doing the job of HR or if it was someone with training and experience.
The examples are a stretch though. I've never heard of anyone losing an interview due to their spelling of nodejs or computer science. I don't buy that's something that happens often enough to spread FUD over.
This is really not an issue. If the company as such a dumb HR department on automatic mode, I’m glad I’ll never be hired there.
But to be honest, I’ve never seen so stupid HR people. Like a lot of people, they have goals to reach and they’ll happily accept an interview from any CV as long as there are computer-related keywords and roughly the required experience.
HRs are one of the most important department when it comes to a company potential shitiness. So I’m glad they have the power to kick you from the hiring process for stupid reasons.
This is a how things are handled at companies that get 500+ applications for each open position and/or are just incompetent. At smaller companies it is very likely that actual humans read a résumé without the keyword matching bullshit.
If you put an idiot in charge of your recruiting, that is the real problem. A perfect AI that would tell you meant "node.js" when your iPhone's autocorrect changed it to "nodule jesus" won't protect you from the jackass running your hiring. There's no protection from being a moron.
Applicant Tracking systems, like Bullhorn, actually do a pretty decent job of fuzzy matching and normalising CV/resumés. Of course there is room for improvement and you need to ‘SEO’ your particular CV, but you really should do that for each job application anyway.
Sadly I think this is a bit idealistic. Even in healthcare, doctors can't agree on abbreviations to put into a patient's records, and sadly the state of EMRs (electronic medical records) is a shambles due to this sort of data impedance mismatch.
Come to think of it, the latter is probably why many of us will always have a job.
If a company is choosing which candidates to interview using grep, it's very likely the rest of their hiring process is similarly broken, and quite possibly other aspects of their corporate culture as well. They're actually doing candidates a favour by giving a clear indicator of how shit they are, much like a candidate who turns up to an interview drunk.
In my company our HR department actually reads all applications.
This is not true for the largest corporations of course. But that is a solvable problem, just provide a standardized form. You can even adapt it to the needs of the industry in question. This is actually a great boon because you can specifically ask for qualifications. There should be enough budget for that at least.
If you company easily discards applications your HR is either incompetent or it just isn't looking for candidates. But even then you try to build a connection to people applying to you. You may meet them later because they tend to work in your industry.
Again, there are special rules for the largest of corporations although I think HR should be manned enough to have a sensible recruitment process.
I recall people copy-pasting the job description and hiding it in their resume by making the text invisible just so that automated software would see a 100% keyword match. Bonus here if you use 2-3 alternate spelling for each buzzword (C++ and c++). Is that no longer done?
Unfortunately the lack of a standardised format hasn’t stopped technology from being developed which automatically reviews CVs before they hit a human. The tech is already out there.
The problem we have currently is that some CVs fail that review because they’re in a format that the tools cannot parse. If you’re CV is strong but you’re seeing fewer responses despite having a prettier CV compared to your peers, then that might be the reason. So from your stance standardisation might actually help the industry rather than harm it.
> The problem we have currently is that some CVs fail that review because they’re in a format that the tools cannot parse.
My company (we have software for responding to applicants in real time... think apply... instant conversation with a human recruiter) did a study of about 1,200 resumes and found that 14% of resumes had inaccuracies that would cause them to be screened out... and about 9% of resumes were in a unparsable format. A lot of the screening we see out there is really bad - mostly text search looking for key phrases like specific colleges, specific employers or specific skills. If you are imagining indexing resumes with elastic search and making queries, that may actually be better than state of the art which is usually something that turns into a SQL query.
Given a lot of these tools cannot even parse CVs with inlined tables, the tools I’m imagining aren’t sophisticated in the slightest. But they are depressingly common.
> Documents like these shouldn't be automatically processed, they should be reviewed by humans.
Human review is fraught with bias. A lot of attention and column inches are devoted to algorithmic bias these days, but let's not confuse that for evidence that human review is perfect or even better. It's just harder to audit.
I don't think anyone is confusing algorithmic bias for evidence that humans are better. It's the other way around -- the argument has always been that algorithms are biased because humans are biased. Having a systemic bias that masquerades as neutral seems worse than having people in the loop.
When submitting my resume and interviewing for jobs, I'd certainly prefer to be subject to: 1) multiple people with different biases, 2) humans that can change their mind and/or recognize their biases, 3) humans that can explain their rationale, and 4) humans that can grasp and weight eccentric and non-conforming experience. I'm not interested in being ranked in a standardized way, because I'm not interested in being a pure cog in wheel.
I also think people who want standardized resumes and automated processing are vastly underestimating the SEO effect it will have on hiring. If resumes are automated, then many people more aggressive than you will game the system to the Nth degree. The same thing that happens to Google and the internet will happen to jobs: the loudest and spammiest will win, and the good content will go mostly unnoticed and unrewarded.
I think resistance to algorithms is about failure modes. If an algorithm becomes standard and you are an edge case then you are in trouble, much the same way you are in trouble today with companies that have automated customer support and make it hard to contact a human being. Losing your email account is annoying but not being able to have a job is worse. Automated systems are designed to avoid mistakes but are bad at correcting mistakes. I would argue that even with humans involved the hiring process is harder to some people, even discounting bias, just because it is more or less standard but at least they have hope in the variance that remains.
I've seen thousands of resumes... it would be nice to have a baseline of requirements for specific positions. An RFC style guide for human parsing . Too many resumes miss what is important.
Knowing life, the end result would be multiple competing standards and weird clans would form along with studies proving that, statistically speaking, standard X people are different from standard Z people.
Edit: Humans are weird. I guess current ecosystem is not as bad as it could be ( even though it does suck ).
" Reducing someone's life history to a list of educational institutions and employers feels robotic ... but that kind of behaviour shouldn't be encouraged."
Reviewing CVs is boring. They should be made easy for the human who is reading. Imagine being a person in a human resources department going through 100s of these... If there was a good portable format, that process becomes far more efficient and interesting.
A good interview after the resume has been accepted would go into precisely what you are advocating for -- the life history and interpersonal context.
Whether or not it should happen doesn't answer the question of why it hasn't happened. The fact is that the very companies that do automated processing of applications probably would welcome resume standardization, so why didn't they do it yet? It's in their power to propose a standard, settle on it, and require it.
I suspect the answer as to why it hasn't happened is simply because most laypeople applicants would find it too difficult to do something like LaTeX or whatever other thing would be necessary to make this a reality, and also, firms in general suck at adopting new technology.
> laypeople applicants would find it too difficult
> to do something like LaTeX or whatever
Those people where a CV and certificates matter, they already construct a CV in the closed system of LinkedIn or Monster's websites. Nobody needs to use LaTeX, there will be websites where anyone can do it, if anything that's an additional business opportunity.
No, I suspect the actual reason is: Nobody will do it unless the big players (LinkedIn etc.) will adopt it. And the big players don't do it because interoperability, and a CV that can be migrated anywhere, isn't in their interests.
A really great resume gets beyond the company names and educational institutions to demonstrate, with concrete numbers, how you improved things at your employer.
Reduced latency by 350 ms
Increased engagement by 18%
Reduced AWS spending by 23%
Scaled from 13 to over 300 virtual machines
That sort of specificity and numerical quantification is what makes a great resume stand out.
I totally agree but we are fighting the tide (ran a resume, linkedin profile, and interview prep shop for 6 years). Everything in the process is being analyzed, parsed and benchmarked using AI, soon to a more granular level. I even came across some interview analysis software which measured the angle of the person's head along with tone and a bunch of other factors!
> I understand that there are real life problems because companies do use automated processing on applications, but that kind of behaviour shouldn't be encouraged.
Ignoring the web, we already do some formatting with the common Education/Work experience trope. We could arrange resumes arbitrarily just to mess with employers, perhaps? They are already pretty easy to parse.
I agree they should be read by humans, but (and I feel most comments are missing that point) a standard format would help with that.
Having been reading resumes a lot lately, I always breathe a sigh of relief when people hand in the standard latex template, with maybe only the colors changed. The structure is immediately familiar and easy to navigate. I could use a client app that presents a hypothetical standard data format in such a way. My colleagues in sales or customer success on the other hand seem to prefer other formatting styles; they could look at their candidates in that way.
Note that I'm not arguing that this it the best presentation possible for a CV. I'm just arguing that having a uniform presentation between CVs reduces the mental burden of filtering through them. (And if you want to get into that, might reduce bias, especially if you don't include the image in the presentation)
Yet, companies use AST tools, which automatically processes the CVs and resumes.
One person that is developing such a tool recommended collage students not to add too much links to their CVs because otherwise, their CV will be flagged as a malicious document.
> I understand that there are real life problems because companies do use automated processing on applications
I'll always remember but never find the post, where someone just hid tons of keywords in a joke resume and got interviews with an alarmingly high success rate.
Reality is, however and in my experience, that more and more HR departments fall for promises of AI snake oil. Virtually every employer I apply with has it's own portal, often times asking you to construct your CV anew on their website (often with drop downs that don't include the actual job title or skill). Preferably with a five minute web session and no submit confirmation. Couple this with companies increasingly not responding at all due to fear of being sued, and you'll spend hours on an application and won't even know whether the non-response is due to you, or due to the website trashing the submission. As for CV parsing, this has gone so far as to a big news site over here posting helpful articles on how to beat the system. "Favorite" tidbit: Submit two CV - one for humans, one for dumb algorithms.
Oh and btw, I'm applying at non-tech companies increasingly. Maybe humans isn't perfect either. I mean, I submit proof of 25 years of high grade enterprise IT experience, just for some HR person to ask me where my certificate as an "IT technian" is: they don't know what any of that "weird stuff" means, but to them I lack a generic "IT technician" certificate in early career. A thing that didn't even exist back then! But the latter may just well be a quirk of my own, bureaucratic country.
But overall, this is a hot mess already, and I have often times had the same thought: Why can't we have some sort interchangeable format, much like "geek code", just for all the possible job titles, universities, locations, and companies; with some format-dependent, but free text fields because one can never catch all of them in a fixed list?
CV or resume analyzed by non-human? Or in an universal format? I hope this will never happen.
How "automatized" way could help there? Unifying the CV/resumes for certain company? Sure, if they are have some form - could be stupid Google Form or more sophisticated like Teamtailor [0] - go ahead. You will have a nice databases or resumes and profiles similar to mini-LinkedIn. While I am fan of automation, I don't see any special incentive here that will profit HR team.
I know hundreds of examples, where CV nor initial job interview didn't exactly has shown how good someone is. Also, I personally had experienced a situation I have completely matched the job offer but after talk during interview, I knew it would be waste of time and mentally it will be a bad choice for next year(s) to join their Acme corp.
Do the JSON databases of applicants will help anyhow to solve the human relation job to find out the best candidate for the position? I am highly skeptical and in my humble opinion it will just "dehumanify" the whole process.
Maybe, it is good if you are looking for some warehouse worker like Amazon? /s
I think you might be underestimating the number of applicants that some companies deal with. I used to work for a firm that did analytics and surveying for the recruitment industry. While applicant to hire ratios in the tech industry might be 10 to 1, in many jobs you can easily be looking at 100 to 1. At those scales companies are using automated CV scanning tools and outsourced humans to whittle them down to a set that hiring managers can actually cope with. This is why many companies insist that you enter all the information into a structured form.
My basic rule of thumb is that the more structured and formal the CV submission process is, the worse the applicant to hire ratio will be, and so the more people you are up against.
For those of us in the tech industry, especially later in our careers, that probably isn't an issue, but in many other jobs, and earlier in your carreer, then it's going to be an issue. I remember having to re-enter my CV into ATSs back in 2001 for the big companies hiring graduates because they had so many to get throught, smaller companies just wanted your CVs.
471 comments
[ 0.24 ms ] story [ 340 ms ] thread1. It lacks diversity (e.g. GenderType is incompatible with German laws, requiring "diverse" gender in official forms, which is definitely not "unisex" defined in the format).
2. It looks very much like a US-centric God Object. It is not clear why DUNS, ISICv4 or NAICS are there, but other identifiers like national ID or SSN are not. It would be better to have a single "identifiers" key-value map instead of them, that would be extensible.
3. Contacts would deserve a dedicated structure and key-value map (why single telephone field? why messenger IDs not there?).
See for example the FoaF ontology https://en.wikipedia.org/wiki/FOAF_(ontology)
I’m highly skeptical of the whole concept and feel like I should have organically come across it by now.
KDE Itinerary and Apple Wallet make use of this. For the KDE Itinerary part, you can read more on this here: https://volkerkrause.eu/ (look at the KDE Itinerary posts)
Banks, research institutes, universities are joined in large consortiums. The German hotel reservation example above is one of the projects that’s getting attention.
In fact, the underlying technologies are all quite mature (triplestore implementations and the RDF, SPARQL, OWL, etc. standards) and used heavily in certain domains, such as biology and language. The standards are designed for easy serialisation and interoperability which is the antithesis of the "modern" walled garden and/or commercial API approach of the big Internet companies, although they also use semantic web tech to a certain extent.
In Europe, you will run into RDF quite often in academia (and research in general). I'm working on the newest version of the Danish WordNet, for example, which is published as RDF and OWL (in the turtle format) and which runs in Apache Jena. Coincidentally, I am actually developing this new version in Clojure which stole the triplestore concept from the semantic web stack and made it a mainstream database type in the Clojure ecosystem (Datomic, Datascript, Datahike, Datalevin, Asami).
Some of these old AAA XHTML 1.1+RDFa websites were well engineered and a joy for both man and machine to behold.
* There are multiple protocols for exchanging EDI documents (AS1, SMTP, FTPS, SFTP...)
* There are multiple standards for the documents themselves (EDIFACT, X12, GS1 XML...)
* There are partner-specific business rules that need to be set up (this partner needs an ASN, that partner requires invoices to be for only one PO each, the other partner can only accept invoices which use their ERP’s internal codes...)
Some of these problems can be papered over with a “VAN” that can translate between standards, but I have yet to see one business send a non-PDF invoice to another without a lot of fuss.
Of course, being a SERIOUS, CORPORATE standard it needs to be overly complex, based on XML and SOAP and WebServices and whatnot: https://www.agenziaentrate.gov.it/portale/documents/20143/23...
And we have to keep filling them out by hand ...
On the other hand, it makes all sense to have a standard "full" CV out of which to select the things to include.
It's not like people are managing dozens/hundreds of candidates with a pen/paper.
There is definitely a need for a standard format.
It shouldn't take too long to parse the desirable skills from the top after that.
Maybe this standard can match against desired/offered compensation brackets as well to get that sweet spot on the bell curve.
Same when choosing a partner: the first thing you use to filter are looks, assuming the risk of filter out a below average looking person that will be your perfect match.
We, of course, delude ourselves into thinking that our method is 100% objective and unbiased. But it never is. Even if you read each one of the 1,000 CVs, you won't read the first one with the same energy you read the 950, and it's likely that you anchor on a given CV that you read among the first 20, and everyone after that is meassured against it.
CVs are by themselves only a proxy to a real person: why don't skip the CV and just interview everyone? Nobody's life fits in two pages.
This page mentions a few more formats: http://microformats.org/wiki/resume-formats
It seems like the problem isn't a lack of standards, but rather a lack of adoption and/or agreement on which standard to follow
At the end of the post, "The larger employers and job-hosting web sites need to get their acts together and work this out".
In a comment right below, the author says that this would be "giving job-seekers more control over their data".
Obviously, the larger employees and job-hosting web sites have no interest in allowing job-seekers more control over their data, quite the opposite, so they won't. And their active cooperation is absolutely required for such a data format to be adopted - so it won't be, at least not in a way that satisfies the author's wish of gaining more control over their data. That's it, it's that simple.
But nobody really uses it for data interchange. I use it to render my resume in new layouts now and then.
Although, it does let style dictate content sometimes (some templates force you to have dates down to the day for job start and end dates, etc.)
As you can imagine, I felt like a clown trying to explain that I didn't have a word document because my resume was generated from a JSON file.
I did, however, have a PDF on a USB drive I always kept on me but they refused to accept USBs out of fear that I was trying to give them a virus. Eventually the lady processing my application gave up and printed the PDF off that was hosted on my website but also scolded me for not having a word doc.
The whole ordeal was pointless anyway since they said they can give me 70% of my rent.
Even if there were a turn-key software platform they all used, which natively supported a standard data format, they'd still find a way to screw it up.
I think over 3k+ people use the new Gist hosting. (In our old hosting we had around 10k resumes. Not including those who by pass the free community hosting)
===
On a personal note, I've loved having my resume in a standard;
- Depending on what type of company/person I am applying to I will change my theme on the fly. (Startup vibes I will make it look hipster, if it's a more formal role I will use a simple black and white theme)
- I use to lose my most recent resume constantly, having it in a Gist called resume.json that I just edit seems to have solved that for me.
- Hopefully one day a standard will get integration adoption so I can just upload my resume.json and not have to fill out the same form fields a hundred times.
I think we wanted JSON at the time because it just relied on less extra packages when working inside of the JS ecosystem.
So yeah, if anyone wants to make a resume.yaml, we only need to add an extra step to convert it to json for them to use any of the tooling.
Why should resumes in Vietnam or China be using English, after all.
In HTTP we have Get, Put, Post, Delete, Patch or Options and not obtenir, mettre, poster, supprimer, corriger, options. Yet the French use the web all the time, as do Chinese or Maui.
It would not be about how the fields are named, but about localised concepts. Maybe there are societies that have some educational concept which western Americans cannot fathom, but which is crucial on a CV there.
This,implies the standard needs to be flexible, extensible and localised. Which kindof defeats having a standard.
So I suppose that parallel suggests it depends on whether there's a realistic expectation of WYSIWYG editors for any kind of universal resume format. Which I think seems incredibly unlikely.
A challenge I see is -- resumes change wildly by industry.
a tech resume is totallly different from an actor's resume.
but I think this complexity can be handled.
I totally support json resumes.
Does the author want the resume to be objectively judged? Maybe not
The reason the author is trapped long enough in meat grinder hiring to notice this "problem" is likely precisely because of some indistinguishable cookie cutter bullet point soup getting them nowhere. If you can capture it in a data structure, it's not a resume!
See also: why isn't there a universal UI for web sites?
Pretty sure a Word file is a data structure.
I would be more likely to locate the problem in increasing numbers of applicants rather than increasing numbers of applications per applicant, but I wouldn't be opposed to seeing a more centralized system go up. American medical schools accept students through such a centralized system. Chinese university admissions isn't centralized to the same degree as American medical schools, but it's still quite centralized, and more formalized.
There are, inevitably, some people who apply to a lot of jobs. Jobs they aren't qualified for. Low intent applications, where the applicant isn't really that interested. Etc.
Even if these are a minority, they apply to a lot of jobs. In any case, standardizing job applications (OP seems to be talking about application forms that are mostly resume-ish fields) just means more of these. More volume, more noise, probably not many more successful hires.
There's kind of the same dynamic on the other side. Most workers don't love the idea of submitting an indexable resume for employers to leaf through.
Sometimes a modicum of friction is helpful.
I agree that a hoop is needed - I recently posted a job on LinkedIn which has some kind of "click to apply" functionality, and it was clear that most of the applicants were just lazy clickers that probably hadn't even read the posting. On the other hand, if you make people jump through pointless procedural hoops, you're screening for people who are ok doing that.
Ask for a cover letter, you get people who actually want the job and hear from them why they are interested.
A cover letter is a kind of writing contest, or maybe it's perceived that way by a candidate. Writing about yourself can be stressful.
I'm not saying not to, ask for a cover letter. Just saying that this stuff is nuanced. Standardisation is great when you want very low friction. Job applications aren't frictionless, probably can't/shouldn't be. Some copy paste fields might be inelegant, but it's not changing the friction equation by much.
The honest truth about why they want the job for most people is that they provide what you need and that they enjoy things like eating, having a roof over their head, and fun things to do with what's left over of the money you pay them.
Coming up with some drivel about how they've been entranced with Version_Five Inc's tech since they were twelve and how it's been their life-long dream to work with their famous Foobar Widget control language is a nice creative writing exercise but it hardly proves anything. If anything, it rewards lying and deceit to get hired quicker.
Many very capable people suck at writing, even if they are truly driven and interested, especially in the computer science field. A cover letter is a great screen for a job that involves a lot of writing, but for most jobs writing is only a side activity that shouldn't be treated as the main objective.
The thing is it's far more than a "modicum" in practice. There was a recent Harvard Business School study out there revealing that these systems are poorly calibrated (and/or these companies don't know how to use them), and as a result millions of candidates are getting rejected for reasons that are either specious (6-month resume "gaps" that may be due to normal life circumstances) or utterly meaningless reasons (like failing to match certain word combinations).
But I agree though that many HR departments would still love to use any tool that narrows the pool (however lossy the filter may be). Whether that is a good thing or bad thing or not is a matter of perspective.
In my view, it puts HR's own value and utility into question.
Yep, that's me taking care of my mother undergoing cancer treatment.
When hire I also like to receive custom resume that helps me to understand better why hire. Again I am totally uninterested tailoring my offering to formalized set of checkboxes concocted by some "industry expert".
Documents like these shouldn't be automatically processed, they should be reviewed by humans. Reducing someone's life history to a list of educational institutions and employers feels robotic even for a software developer's mindset.
I understand that there are real life problems because companies do use automated processing on applications, but that kind of behaviour shouldn't be encouraged.
I’d wager in many cases they aren’t even aware of an failure to process a resume occurring.
But I think an official format for CVs might not work so well because CVs in different fields aren't formatted in the same way. A programmer isn't going to put forward the same kinds of things as a musician or a professor in psychology.
If resumes could be standardized and easily searchable for information it would help find candidates much more accurately and quickly without having to read.
Even better would be if there was a verification standard where you could get a blue checkmark on your resume meaning all the information is accurate and not made up bullshit, then you could limit searches only to verified resumes and do less investigation or third party background checking.
It pushes industries forward.
I see how that can be useful from the recruiting side of the table, but fail to see how the candidate benefits.
The key point here, I guess, is that a lot of people have told me my resume is in the wrong format, but I still got the job. What's the right format for one employer is the wrong format for another employer. So there is no way to win. I now have no clue how to write a resume, but it also doesn't seem to matter.
(Still bitter about how few people ever look at my Github, though. People put that as a custom field in their automated system, and then the interviewers have nothing to say about my projects!)
We have a "verification standard" in the form of diplomas and certificates and neither of those have fixed the hiring issue so far.
If you want to skip background checking, just make people bring their proof of certification with them to the job interview. Of course, those certifications barely mean anything in most fields of work, but it's an exact equivalent of the blue checkmark system you propose.
I get that hiring is hard, but that's why hiring managers exist in the first place. If we used a nicely standardised, automatically validated system, all recruiters and hiring managers would be out of a job. Why pay someone to do that stuff when you could pay a cheap machine learned AI to fill a list of requirements for you?
Effective way to select for only young employees, I suppose.
Way, way more than a matter of parsing a bunch of standardized resume documents.
That's very nice to think about but in the real world if people don't match some exact keywords they get thrown away. HR is looking for "node.js" but you wrote it like "nodejs"? Though luck pal, bye. HR wants a "computer scientist" but you have a degree in "computer science"? Same, automatically discarded.
If anything, a standard format would at least allow people to classify themselves/others correctly and without ambiguity.
It's random, and not based on any job-relevant differences, for sure. But hiring is always going to be pretty random so long as we're pretty bad at predicting who'll be good at any given job. So long as there's no systemic bias, it's fine, and selection by idiosyncratic spelling is at least much less biased than a lot of other irrelevant criteria that get used in hiring.
I'm sure this happened somewhere at some point in time, but IMO it's becoming a myth blown out of proportion.
I've never actually worked at a tech company that had any sort of automatic resume filtering software. As a hiring manager, we didn't even have HR pre-screening resumes for us. If we did have recruiters, they were competent and incentivized to find good candidates (e.g. wouldn't be unnecessarily rejecting good candidates because they were missing exact spellings of keywords)
It would seem ludicrous in 2022 and this hiring crunch for a tech company (one that you'd actually want to work for) to arbitrarily filter out resumes based on incomplete keyword searching.
That said, I have heard this often enough that some people must do it. I can't imagine why though, as it takes only a few seconds for a human to scan a resume. Even at 10 seconds per resume (a conservative estimate), that means the human could screen 360 resumes per hour.
Based on experience at firms with on the order of 1:1 applicants to hires up to 10,000:1 applicants to hires, I’d agree most such tools are not present at most companies, as they tend to be used from ratios of 100:1 and up.
On the other hand, while most companies might not have or use such tools, the more jobs a company has, the more likely they are to have some form of selection or filtering in the pipeline.
But none of the ATSs really collect great data on how many of their customers use them in their core processes. From observing actual, heavy duty recruiter and HR usage at Ladders, I suspect it is not very many.
These are people people after all, and they are not great at writing REGEX style queries to find only the precisely correct resumes they are looking for. You also need to remember that they are frequently working on novel searches each week, and wouldn’t have the confidence or the capability to craft such searches.
Tech companies doing recruiting of any meaningful volume are indeed generally using searches or matches of some kind at the start of the funnel.
A standard format wouldn't solve the problems you described. There's no comprehensive list of frameworks and programming languages, let alone for types of experience outside computer science. The data would be structured, but HR would still throw out the bits of data that need to be filled in manually.
Minor anecdote incoming. I applied to a company A. It was a fit. I think I used the right keywords. Zero response. I talked with my buddy there. She talked with hiring manager. I got hired by company A. I later found out HR was unhappy, because it undermined their 'process' - such as it is.
There just has to be a better way to do this when compared to things as they are now.
Fundamentally it does kind of suck to have to put up people in your environment who are "unhappy" with you, or the fact that you snuck in there -- not because you don't belong there, but because of their own basic ignorance as to how things actually work.
Not that you can't grit your teeth and live with it - it's a job after all, and that's what you get paid for. But still, it does kind of suck.
Any standardized process will lack fidelity and excellence in specific cases. Or maybe even in all cases, when the company has made the trade off for cost or speed over quality.
But that’s what big company life is like, and that’s why they have standardized processes. Because even with the errors, in aggregate, it works out better for the business.
Wouldn't those issues be more likely to occur as a result of automated processing? I imagine that most humans who deal with technical hiring would guess that "node.js" and "nodejs" are probably the same thing.
http://web.archive.org/web/20171002221201/http://www.jerkcit...
Originally published in October 2002 (almost 20 years ago from when this comment was made, for you future readers and time travellers).
“Do not send in a job application before you have an actual human being who has expressed enthusiasm for reading or forwarding it.”[0]
I believe this is sound advice, and the pitfalls you mentioned was probably the reasons that he gave this advice.
[0] https://twitter.com/patio11/status/770807498802540544
Networking and reaching out to people at interesting companies is another way.
You will probably still have to write a cover letter and send in your résumé. But if you have someone on the inside who will mention to the hiring manager that, ”hey, you should check out Jane’s application” it will probably make a great deal of difference compared to just applying and hoping for the best.
And please folks, remember that FAANG is not the only game in town. There are lots and lots and lots of other companies out there, and many of them also pay well since they need to compete with FAANG to at least some extent.
I definitely think it’s likely that most people on hn don’t work for companies that do this, but maybe hn just skews toward certain kinds of tech jobs.
I think the fact it’s low cost to avoid and high cost if it occurs keeps most people just doing it regardless of whether it’s true or not
For external candidates, we see less than 30% of all resumes. As with all things, 80% of the rest are junk and another 20% are potentially viable candidates.
It would seem most of the over-confident “it’s a myth” comments in this thread are from folks not familiar with typical processes for hiring in the 100s to 1000s.
We consider ourselves lucky to get 50 résumés, and if even one isn't obviously disqualified that is a miracle
This isn't the "real world" you're referring to here -- it's the fin-de-big-tech bubble world where everyone and their dog is not only sold, smack-down drunk on the idea that Algorithms and Data are the Solution to Your Problems. They aren't of course -- it's just a giant hornswoggle. This "world" needs to end, and it needs to end soon.
My belief is that it will. But until it does, companies that run crappy ATSs (like there's any other kind) deserve the buzzword-gurgling, keyword-dropping candidate pool get. And developers who get "rejected" by these companies should be grateful for the sublime gift of this rejection -- and for the opportunity to laser focus on companies and teams that use their heads to hire, rather than a fleet of bots.
Snark aside - hiring managers that actually read resumes (and yes, actually at least skim each and every one -- really it ain't that hard) are golden to work for (other factors being equal). Really, you don't want to waste your time with companies that have drunken the ATS kool-aid. Really you don't.
That said, however:
HR is looking for "node.js" but you wrote it like "nodejs"? Though luck pal, bye.
It's "Node.js" with a dot. If you spell it "nodejs", that's a serious red flag and you shouldn't be surprised if it gets you automatically discarded -- even by a human reviewer.
That's what resumes used to be, after all -- a kind of a take-home test where you really do have enough time (and perfect knowledge) to get everything 100 percent right. And also a test of your awareness of the fact that, yes, in critical business communications at least -- this level of correctness does matter, and it matters a lot.
So if Node.js has that many more hits than nodejs, it's not surprising that some people see it as canonical.
Hearing the context of this conversation, though, I can see that "nodejs" in your resume is a signal to senior peers within that community--others who are "in the know". That may be a useful filter.
The only thing this filters for is people who have been looking for a job for a long time and/or are people focused on presentation instead of content, good luck with that.
We’ve hired incredible engineers across that have had typos and major resume mistakes many times and they had no communications problems on the job.
A resume is a background/interests check for the hiring manager. Giving them a spell check is about as useful as measuring the ratio of times the letter ‘a’ appears relative to ‘e’.
> critical business communications at least -- this level of correctness does matter, and it matters a lot.
This sounds like someone who is unable to contextualize when that would be relevant. Unless you’re writing user documentation there are zero times in an engineer’s day-to-day communications where writing nodejs instead of “Node.js” would have any impact at all.
I know I've rejected resumes for tech writing positions that contained spelling errors. I would hope a tech writer would at least use a spell chekker.
I'll leave the matter of whether one should be expect resumes to be at least reasonably correct (in terms of spelling and grammar) or not - or what the threshold for "reasonably" should be on this front - to the side for now. It's obviously a question of taste, to a certain degree.
But sorry - what you're saying about the ratio of 'a' to 'e' makes no sense at all.
And on this front:
Unless you’re writing user documentation
I don't know what planet you live on -- but on my planet, good engineers write "user documentation" (in the sense of READMEs, docstrings, etc) and participate in mailing lists, do conference presentations, etc, all the time. And yes, while nobody is 100 percent perfect, consistently making a large number of gaffes (or more to the point: conveying an attitude that these things just don't matter; or that you're so hot that they shouldn't matter for you) in these contexts -- particularly when there there is plenty of time available to edit and review -- is definitely seen as a red flag, both in and outside the company.
For any company you'd want to work for, that is.
These days I think I work at the best team I ever worked on and it seems I'm in the top two in spelling, it is almost as if the rest if the team is mildly dyslexic or doesn't care.
But they are nice to work with, smart and they make good code.
I'll take them 49 times out of 50 over some perfect in code and writing but nasty in person ex colleagues.
Even if I really really really need a specialist on hight performance <whatever it was, I can't add a dead giveaway here> I'd still try to look elsewhere first.
Edit:
... and those two ex colleagues? Their code was good but it wasn't perfect either.
Sorry I confused you. Let me be clear, it’s not relevant at all to job performance. I would take someone with terrible spelling and punctuation who actually writes documentation over someone who writes half as much because they are obsessing over details like that.
> in the sense of READMEs, docstrings, et
These are generally for other developers. Developers do not get confused when they see “NodeJs” instead of “Node.js”. Do you know what point you were trying to make?
> is definitely seen as a red flag, both in and outside the company.
> For any company you'd want to work for, that is.
Not sure what kind of company you’re talking about, but it’s definitely not a tech company. Maybe Wall Street or something a little slower paced?
I've seen my fair share of resumes and hired my fair share of people, and I've come to the conclusion that sloppy, error-ridden, CVs are a sign of a lack of attention to detail.
Here's a candidate with all the time in the world, and huge motivation, and they still can't complete this task correctly. It's not a good look.
I'm not saying a single error I'd a deal-breaker, but it puts the applicant on the back foot. No doubt many of my rejects are quality engineers, but engineering, and software development are all about details.
You are woefully out of touch with the job market. And I’m saying this as someone who worked as a hiring manager at one of the most applied to tech companies in the world.
Edit: oh, and also US state governments. there's a reason they're known for being incompetent, people job hop within months and get keywords on their resumes and nothing ever gets done
Oh, really!? I'll be very glad NOT to work at a company that throws aways candidates because of such triviliaties.
In some software, say, nuclear warhead tracking software, you want those people.
I write it as nodejs because that's how I got used to it on the mailing lists and several places that I frequent.
If a "team" decides not to hire me because of the latter, then ... thanks, I guess?
But at this point we're getting into the territory of what's seen as a positive in some niche communities may not be seen as a positive signal outside that community. And in business communications, what masters most is knowing who your audience is, and how your message is received.
It’s ego.
What is the résumé’s job? To convey experience and work in a positive light. A small typo that may incidentally be the correct spelling shouldn’t derail that if the goal is to hire the most talented people.
To my defense, I was going by my Bayesian prior for certain language names for which one could be 100 percent sure there was no alternate accepted spelling or capitalization.
So if one put JAVA or PERL on their resume, that would generate a definite "pass" signal.
Or even if it's just 1 or 2. It's infinitely better to hire someone who is in fact really solid at what they say they know, than someone who tries to spam you with every random keyword they've been exposed to in the hope that you won't drill down and ask them any hard questions about most of them.
That, and skill listings that don't pass the "apples and orange" test. For example:
The former 3 are full-scale programming languages, but the ones that come after obviously are not.Know yourself, and know your audience.
It is never advice toward hiring manager. It was always criticized as not smart and leading to many false rejects.
Yikes. “It’s not JavaScript it’s Ecmascript! What a RED FLAG.”
Not comparable.
It really does require an eye for detail (in places where it matters) to a degree that comes epsilon close to being "pedantic" in the annoying sense -- without quite going over that line.
Making such „mistakes“ actually shows quite a bit about the counterpart starting with correctness and precise communication.
Back then it drove me nuts when someone wrote „advanced in AngularJS 2“ - and it kept coming!
Mine certainly isn't, it's like me. Worked out so far.
But then again, I wouldn't want to work for most companies.
Pecunia olet!
But my CV also shows part of my character, mainly the fun part. If that is deemed unprofessional by a company, I immediately know they are no good fit for me.
The thing is: As with every relationship, they are two way streets. I am not a beggar as a potential employee, neither am I king. We see eye to eye or not at all.
At least fixing stupid mistakes in your resume is well within your control, and (unlike cramming for their tests) doesn't take hours and hours of your time.
But to reject a candidate just because they're not the same kind of pedant that I am would be quite unreasonable IMHO. As long as the whole thing isn't littered with typos or formatting issues, that's just something I wouldn't pay too much attention to. Not everyone needs to be the person to dot every i and cross every t.
Also, I really want to see the HR department that can confidently reject or even mark down a candidate when most of the writing I've seen from most people in any company I've ever been (including from HR departments, PMs or executives) has been "meh" at best, and full of errors or at least awkward language more often than not. For better or worse, writing skills are not necessarily something we tend to select for in the industry.
But whatever. It's your life. You do you, and I'll do me.
https://www.reddit.com/r/javascript/comments/8d0bg2/oracle_o...
From my experience, however, it's the native English speakers who spel the worse.
Besides, I've noticed that ESL speakers make different kinds of mistakes than sloppy native speakers do.
All of the lines below have red squiggly lines under them in my browser's text area:
- Qt Quick
- Datalog
- Datomic
- TensorFlow
> I've noticed that ESL speakers make different kinds of mistakes than sloppy native speakers do.
I've never seen a non-native speaker writing "would of" or using "they're" instead of "their". On the other hand I can't count how many times I've skipped an article or used a wrong one somewhere (since my native language doesn't use them at all).
She has internal word concepts. When the English words fit cleanly into her existing concepts it's easy. When two English words map into one word concept it's considerably harder, it took her a long time to correctly separate turn on from open (and turn off from close), but at least any given thing always goes the same way. In situations where that doesn't even work she will make frequent errors even to this day--do not trust her use of gender words.
It’s a document sent to tons of people. Revised by tons of hiring managers. Can’t you spend 5 minute looking at a dictionary? If you cant be bothered to spend 5 minutes researching this for an important document, what makes me think that you will spend 5 minutes trying to be good at your job. Lack of care on your resume is lack of care on lots of other things.
There was even one time they forgot to give me the resume... that was probably the funniest interview ever. They got the job too.
Bullshit. There are many things person may care about and many things they may not. If this position is unrelated to professional writing, it's just irrelevant. Are you questioning their abstract ability to "care" and trying to translate spelling care to job care? Bad news, it doesn't translate (aside from the fact that ideal perfectionists can't get shit done ever; your best candidate still figures out the ideal form of his resume and is too anxious to click "publish").
Perfectly, no. But in formal communications, they need to show that they're at least making an effort to get things reasonably correct. And if it matters enough, you can always find a native-speaking friend to review your resume or cover letter for you.
(Source: been there, done that, applying for jobs in other countries).
Of course, it is also reasonable to be forgiving on resume errors. It depends on the culture and team you're trying to build.
Literally every time I have submitted an (academic) paper at least one US reviewer had to point out I made several spelling mistakes when I actually used the British spelling (consistently) instead of the US spelling.
Sorry my man but it is called "English", I think I am allowed to spell things like they do in England.
Agree. And yet, they are still the ones that mix up "your" and "you're", "then" and "them" and "than"
And for people who want to work in startup land, it’s an ideal.
Once a company gets past 40 or so people, though, ATS adoption increases. And 150+ employees without an ATS are rare. It’s just that the volume of candidates and interview processes that need to be managed require there be some software to manage the workflow.
on edit: before people come in with not wanting to work for a company that would make such a stupid mistake, first off I believe mistakes sometimes happen, and second Henry Thompson leader of the XML Schema working group, had the subtitle of his LinkedIn page saying something like he was an expert in XLM for at least a year (I should really have taken a screen shot) so I mean I think it would just as silly to not work someplace because they made a mistake as it would for them to screen you out for making a mistake.
https://news.ycombinator.com/item?id=15187874
- resume keyword scan (automated or otherwise)
- leetcode interviews
The former I think you’re spot-on. If you’re serious about hiring, some qualified to do so needs to be reading resumes.
The latter is complicated. I tend to agree that the industry has over-pivoted to “dynamic programming whiteboard puzzle” as the LSAT of high-paying software work. But while most seem to agree that leetcode interviews suck in tons of ways, something that is unambiguously superior seems to be an open problem.
And some algorithms and data structure testing is important for hard software work. Even choosing the right library requires some basics. Is every SWE at Amazon implementing CS papers every month? No, clearly not. But unless machine resources are free in your domain, brute-force doesn’t always work.
FANG people do in fact get elitist about their CMU educations. But there’s another crowd that throws the baby out with the bath water: “I solve business problems, not reverse linked lists.” Well, what kind of business problems don’t involve some computation?
It’s a balance that our industry seems to still be trying to find.
Here is your pink slip. Don't call us
Best example of someone I wouldn't want to work for ever
perfect filter, both ways
Two ways of three. Some innocent person may accidentally put Node.js in their resume and get hired.
Haha and then they scream about labour shortage )
I get hundreds posting into some of the roles I’ve hired for and I’ve got to look at every single one in order to find a possible match. It generally takes me about 30 seconds to a minute to assess a resume. A kernel of interest stands out quickly or a smoke screen of buzz words tells me it’s the wrong candidate.
Algorithms aren’t a substitute for an active hiring manager being really interested in finding the right candidate to join their team.
He said he looked at about 1000 resumes a week. After a while, you can tell in 2 seconds if it is worth pursuing or not.
Except it doesn't, because a lot of things in life doesn't fit in that boxes that well.
You might not have experience in node.js, but in JS and some of the node.js-alike server side VMs.
Even just a "degree in computer science" can mean many things. Furthermore you might have something which for the position is "equivalent", like a "master of science in cognitive systems with a specialization on AI" which contained a lot of CS courses, and maybe also some pr-axis experience. Or you just started working as a programming free lancer with 16. Or ...
Furthermore a resume is quite individualistic, beyond just the raw content it contains. How that content is represented can sometimes give you hints, about what kind of questions you want to ask during an interview.
I have seen a bunch of standardized forms for resumes some companies opted into, they _always_ caused endless problems.
Fixing that would require not only a universal data format but also a universal taxonomy of terms to describe skills. That would need everyone to agree on how to describe what they do. Easy enough when you're describing programming languages, but effectively impossible for any softer skills. This is precisely why you need the human aspect.
The way to fix the issue you describe is to educate HR people, not to try to apply technology to a non-tech problem.
Resumes aren't just a list of skills and work history (education isn't even on my resume anymore), it's also a chance to let a small amount of your personality show through so you stand out against the pile.
Of course, if you're submitting to a big corp, you'll be fighting against AI reading your resume first, so, write for that if that's how you roll. I usually only apply to smaller companies that apparently end up as bigger companies by the time I move on.
I like that so much that I dare not copycat it.
in the house == domesticated
house approxEqual box
https://www.linkedin.com/mwlite/in/mikkelpaulson
"Keeps things running. Puts out fires."
Served me well on the resume, probably for quite similar reasons.
I wish it was easier to do. At my first job after university (not my first job altogether) we initially had great rapport between "DCOps" and HR. To be honest, our group was rather... irreverent, which made it even more interesting that we had such great rapport (we had all the stereotypes going for us, including female-only HR team and basement-dwelling male-only sysadmins).
But thanks to great cooperation, HR would do basic filtering on things they actually checked with us about, then forwarded us the CVs unmodified to give our own opinion before deciding whether to go with it or not.
Unfortunately some time later head of HR ended up let go due to some conflict with C-level I believe, and things slowly reverted to the mean :(
A regular HR can only do CTRL-F CTRL-V efficiently in their PDF viewer, when automated system would likely test for /\bnode(?:\.?js)?\b/i, because it's a result of a development cycle, which usually includes some field analysis. HRs would LOVE to select something like "where has($node) and years($redux) >= 2 order by years($php) desc". They are not idiots, they just have piles of data and nonsensical tools. Of course they optimize for nonsense.
A human could search for different variations too, but they are average in average and have no the expertise to install a regex-able viewer. It's stupid little things like this that prevent them from doing their job better, not some systemic issue of their own.
Disagree.
Recruiters, hiring managers, and specialized software is all I see.
The examples are a stretch though. I've never heard of anyone losing an interview due to their spelling of nodejs or computer science. I don't buy that's something that happens often enough to spread FUD over.
But to be honest, I’ve never seen so stupid HR people. Like a lot of people, they have goals to reach and they’ll happily accept an interview from any CV as long as there are computer-related keywords and roughly the required experience.
HRs are one of the most important department when it comes to a company potential shitiness. So I’m glad they have the power to kick you from the hiring process for stupid reasons.
Come to think of it, the latter is probably why many of us will always have a job.
If a company is choosing which candidates to interview using grep, it's very likely the rest of their hiring process is similarly broken, and quite possibly other aspects of their corporate culture as well. They're actually doing candidates a favour by giving a clear indicator of how shit they are, much like a candidate who turns up to an interview drunk.
This is not true for the largest corporations of course. But that is a solvable problem, just provide a standardized form. You can even adapt it to the needs of the industry in question. This is actually a great boon because you can specifically ask for qualifications. There should be enough budget for that at least.
If you company easily discards applications your HR is either incompetent or it just isn't looking for candidates. But even then you try to build a connection to people applying to you. You may meet them later because they tend to work in your industry.
Again, there are special rules for the largest of corporations although I think HR should be manned enough to have a sensible recruitment process.
The problem we have currently is that some CVs fail that review because they’re in a format that the tools cannot parse. If you’re CV is strong but you’re seeing fewer responses despite having a prettier CV compared to your peers, then that might be the reason. So from your stance standardisation might actually help the industry rather than harm it.
My company (we have software for responding to applicants in real time... think apply... instant conversation with a human recruiter) did a study of about 1,200 resumes and found that 14% of resumes had inaccuracies that would cause them to be screened out... and about 9% of resumes were in a unparsable format. A lot of the screening we see out there is really bad - mostly text search looking for key phrases like specific colleges, specific employers or specific skills. If you are imagining indexing resumes with elastic search and making queries, that may actually be better than state of the art which is usually something that turns into a SQL query.
14% with incomplete information also sounds right.
And I also concur with you on the screening being pretty basic, not ES level of sophistication at all.
Human review is fraught with bias. A lot of attention and column inches are devoted to algorithmic bias these days, but let's not confuse that for evidence that human review is perfect or even better. It's just harder to audit.
When submitting my resume and interviewing for jobs, I'd certainly prefer to be subject to: 1) multiple people with different biases, 2) humans that can change their mind and/or recognize their biases, 3) humans that can explain their rationale, and 4) humans that can grasp and weight eccentric and non-conforming experience. I'm not interested in being ranked in a standardized way, because I'm not interested in being a pure cog in wheel.
I also think people who want standardized resumes and automated processing are vastly underestimating the SEO effect it will have on hiring. If resumes are automated, then many people more aggressive than you will game the system to the Nth degree. The same thing that happens to Google and the internet will happen to jobs: the loudest and spammiest will win, and the good content will go mostly unnoticed and unrewarded.
Edit: Humans are weird. I guess current ecosystem is not as bad as it could be ( even though it does suck ).
Reviewing CVs is boring. They should be made easy for the human who is reading. Imagine being a person in a human resources department going through 100s of these... If there was a good portable format, that process becomes far more efficient and interesting.
A good interview after the resume has been accepted would go into precisely what you are advocating for -- the life history and interpersonal context.
I suspect the answer as to why it hasn't happened is simply because most laypeople applicants would find it too difficult to do something like LaTeX or whatever other thing would be necessary to make this a reality, and also, firms in general suck at adopting new technology.
Those people where a CV and certificates matter, they already construct a CV in the closed system of LinkedIn or Monster's websites. Nobody needs to use LaTeX, there will be websites where anyone can do it, if anything that's an additional business opportunity.
No, I suspect the actual reason is: Nobody will do it unless the big players (LinkedIn etc.) will adopt it. And the big players don't do it because interoperability, and a CV that can be migrated anywhere, isn't in their interests.
Reduced latency by 350 ms Increased engagement by 18% Reduced AWS spending by 23% Scaled from 13 to over 300 virtual machines
That sort of specificity and numerical quantification is what makes a great resume stand out.
that's a flawed view imo.
there could be plenty of catch-all sections to maintain "descriptions"
resume content expectations are largely driven by industry.
i've not heard of someone reading resume more than 10s.
the whole job posting industry dug themselves into a meaningless, needless task: parsing resume pdfs. heck, some even use "AI" for it.
Ignoring the web, we already do some formatting with the common Education/Work experience trope. We could arrange resumes arbitrarily just to mess with employers, perhaps? They are already pretty easy to parse.
Having been reading resumes a lot lately, I always breathe a sigh of relief when people hand in the standard latex template, with maybe only the colors changed. The structure is immediately familiar and easy to navigate. I could use a client app that presents a hypothetical standard data format in such a way. My colleagues in sales or customer success on the other hand seem to prefer other formatting styles; they could look at their candidates in that way.
Note that I'm not arguing that this it the best presentation possible for a CV. I'm just arguing that having a uniform presentation between CVs reduces the mental burden of filtering through them. (And if you want to get into that, might reduce bias, especially if you don't include the image in the presentation)
One person that is developing such a tool recommended collage students not to add too much links to their CVs because otherwise, their CV will be flagged as a malicious document.
I'll always remember but never find the post, where someone just hid tons of keywords in a joke resume and got interviews with an alarmingly high success rate.
It might be this one?
Reality is, however and in my experience, that more and more HR departments fall for promises of AI snake oil. Virtually every employer I apply with has it's own portal, often times asking you to construct your CV anew on their website (often with drop downs that don't include the actual job title or skill). Preferably with a five minute web session and no submit confirmation. Couple this with companies increasingly not responding at all due to fear of being sued, and you'll spend hours on an application and won't even know whether the non-response is due to you, or due to the website trashing the submission. As for CV parsing, this has gone so far as to a big news site over here posting helpful articles on how to beat the system. "Favorite" tidbit: Submit two CV - one for humans, one for dumb algorithms.
Oh and btw, I'm applying at non-tech companies increasingly. Maybe humans isn't perfect either. I mean, I submit proof of 25 years of high grade enterprise IT experience, just for some HR person to ask me where my certificate as an "IT technian" is: they don't know what any of that "weird stuff" means, but to them I lack a generic "IT technician" certificate in early career. A thing that didn't even exist back then! But the latter may just well be a quirk of my own, bureaucratic country.
But overall, this is a hot mess already, and I have often times had the same thought: Why can't we have some sort interchangeable format, much like "geek code", just for all the possible job titles, universities, locations, and companies; with some format-dependent, but free text fields because one can never catch all of them in a fixed list?
How "automatized" way could help there? Unifying the CV/resumes for certain company? Sure, if they are have some form - could be stupid Google Form or more sophisticated like Teamtailor [0] - go ahead. You will have a nice databases or resumes and profiles similar to mini-LinkedIn. While I am fan of automation, I don't see any special incentive here that will profit HR team.
I know hundreds of examples, where CV nor initial job interview didn't exactly has shown how good someone is. Also, I personally had experienced a situation I have completely matched the job offer but after talk during interview, I knew it would be waste of time and mentally it will be a bad choice for next year(s) to join their Acme corp.
Do the JSON databases of applicants will help anyhow to solve the human relation job to find out the best candidate for the position? I am highly skeptical and in my humble opinion it will just "dehumanify" the whole process.
Maybe, it is good if you are looking for some warehouse worker like Amazon? /s
[0]: https://www.teamtailor.com/en/
My basic rule of thumb is that the more structured and formal the CV submission process is, the worse the applicant to hire ratio will be, and so the more people you are up against.
For those of us in the tech industry, especially later in our careers, that probably isn't an issue, but in many other jobs, and earlier in your carreer, then it's going to be an issue. I remember having to re-enter my CV into ATSs back in 2001 for the big companies hiring graduates because they had so many to get throught, smaller companies just wanted your CVs.