I don't like the requirement to sign up for Github in order to sign up for your thing. Seems like an unnecessary hurdle considering how few real-world developers actually have profiles there.
Worse, I do actually have a Github account, but I'm not going to tell you about it in order to log in and learn what this thing does, given that I have no idea what you plan to do with it. I can imagine a scenario where I'll discover a page on the internet proudly proclaiming that "Jason Is Looking For A Job!" when the reality is simply that "Jason is curious what dimensions go into that shape thing".
Give me a user/pass. Or better still, no requirement at all to dig in further. Then I'll dig.
My favorite part about the GitHub login is that since I keep my dotfiles on GitHub,these type of applications want to know my skill level and job preference based on the breakdown of languages in my profile.
I have yet to see a senior level .vimrc architect job in the wild.
"few real-world developers actually have profiles there"
Is that really true? I work for a mega-corp and I still put as much work as I can on GitHub. You always have side projects you can put up there as well.
I interviewed at Amazon 2 years ago and they wanted to do a real time coding session, but had no way of easily sharing code. I said I'd just post it in a GitHub Gist and read out the URL and he was dumbfounded and had never heard of this before. So yeah, _lots_ of developers don't have GitHub profiles or even know what it is.
It's a good (open discussion, latest topics and trends, expanding the palette with exposure of ideas) and bad thing (secret/slower paced work environments, unexposed devs, rockstars that don't need it, etc).
I work for a megacorp. I have a few unfinished personal projects on github but everything else if confidential. You're not going to find out much about me based on my github account. However, it seems to be a requirement for a lot of jobs these days, especially in startup landia. So if you can't share things you do at work then you have to spend all your free time doing OSS or side projects that you can share on github or no one believes you can actually do what you say you can.
> So if you can't share things you do at work then you have to spend all your free time doing OSS or side projects that you can share on github or no one believes you can actually do what you say you can.
Nah, just keep on looking for work at a place that is willing to invest time in evaluating you.
I work for one too and it's hard to even discuss Git itself with someone here; most people just know SVN. The stack in my industry is very stagnant and I think a lot of people just learned what they had to when they were still young and became set in their ways. They have families now and coding has become a 9-to-5 chore; they're not interested in keeping a portfolio up to date or even trying anything new.
I'd presume most people who come to HN are interested in evolving their skill set. The common worn-out office drone doesn't visit HN.
I have always used git, privately (eg: Bitbucket). And if I want to run an open source project, I can very well do that with Bitbucket too. I don't know why employers ask to see a Github profile.
It's really true that most developers don't have Github accounts. There's also not a strong correlation between sophistication and having a Github account.
Most developers know about Github, of course, and have used it. You know, to get stuff.
I suspect so. I have a profile, but there are only some trivial programs there. Of developers that I am contact with these days, maybe at most 10 percent of them have put profiles there.
| few real-world developers actually have profiles there.
What is the actual ratio of developers on github to not on github? All the developers where I work are on github, as well as the developers who I know personally. And if not on github, then bitbucket.
I have a Github and I actually have a few public projects but the vast amount of my work is private.
Funny thing is that even if you create side-projects it's unlikely that you'll actually want to make them public. For example, say you want to create a Saas or bootstrap a small startup. The idea of putting all of your execution online for all to see (or borrow) is super scary.
For me it's about potential upside. If I take time off work to work on my own projects and I expend 10 hours a day in creating them, am I doing this in order to lock down a better job in a couple of months, or am I doing this because I would like to some day have personally crafted something which creates value for others and allows me to live freely?
I think you'd have to be crazy to want to put huge projects online for all to see (particularly as they can often require stigmatising relaxation of processes which don't make as much sense as an individual in order to be able to handle the workload as a sole engineer.)
Of course there are those that like to build development tools and libraries for other engineers. That's a priesthood I've not really entered properly - it's hard to create something which isn't just yet another me-too framework, but very respectable if you are able to do this (top ~1%).
90% of Stack users do not have Github accounts. That was my intel from an insider source from Stack. Got no proof beyond that but intuitively feels about right.
NB: lot of MS guys on Stack - maybe a correlation there
We thought long and hard about what type of marketplace we wanted to create. We ultimately felt that there was a requirement to gate the marketplace on both sides to prevent bad actors from parasitizing the site (see LinkedIn). Engineers are in such high demand, that they go, inevitably recruitment agents and others will follow.
Github isn't perfect, but we our use of it is primarily as a signal of the type of people we want to attract (developers only) and as an additional admin hurdle for bad actors to have to go through.
So, yeah, we took a risk. Higher friction at the both ends (employers are curated, again to stop agents), but our aim is to create a high signal, zero noise matching service for engineers and this was the way to do it.
Our plan is to add further login options - bitbucket, stack et al. v2.
Hope that explains some thinking about the sign up flow - feel free to ping me any questions!
Along those lines, is a cross-login of any kind really ever a filter for anything? Seems like it's generally more either a huge hassle or something to make seamless login much easier for people who don't care about using their facebook login for everything.
Shouldn't the content/workflow/UX of your site be what attracts certain kinds of users and disincentivizes others?
You make a very good point about content! Its true, that the content on the platform will determine whether the users get value from it. The caveat missing from your model is there are different types of users, with different, often opposing agendas, who might impact that UX.
Suppose for instance, we have a website where the content is generally great tech opportunities with cool companies.
Who's interested in that?
Engineers interested in job discovery might be and that is precisely our target audience. However, there are others who are also interested in that sort of information - recruiters, for instance, looking for leads. Now this is not the community we are trying to serve. In fact, we know that this demographic has the potential to seriously damage the UX of the employers of the others side of the platform (imagine if you signed up as a hiring manager to Workshape and it was advertised as direct employers only - and all you got in return was recruiters hassling you to give them a mandate to recruit for you...). So, we put into place mechanisms which firstly signal who our audience is, and secondly create friction to get in.
We understand that this is at some cost to rapid scale. But we don't value a platform which scales quickly, then equally quickly becomes toxic with the effluent of people who shouldn't be there.
Hope that makes sense! Keep the questions coming - feedback is welcome and necessary.
We know Github isn't a secure gate, but we also know that it is at least a gate - an extra admin layer to go through.
The analogy I'd use is similar to street crime. Your bike isn't secure because you put a D-lock on it. You're just making a little bit harder for the criminal to get it. A motivated bike thief? Yeah, he'll get your bike! But your casual criminal might just pass on rather than invest the time.
What we do want to avoid is to become a LinkedIn-like open platform, where recruiter parasitization is endemic. Our plan for a high signal, low noise platform is absolutely dependent on being agency free
Hope that answers your question. Keep them coming - all thoughts welcome!
Interesting, but I don't see why I should give them a preferred job title. Interests, skills, job level -- of course, that's how they match people. But preferred job title and description? I don't want to end up in a catalog, I just want to see positions.
Upfront an engineer might specify that they'd like to spend 60% of their time on Data Science and 20% in Architecture but this doesn't mean that they are skilled appropriately for this.
What about companies that want to hire people to fill roles that require very specific experiences and education?
Is this app assuming that we when asked to describe what we want to do, we won't describe anything too far from what we already do?
Is the market on Engineers sides so much that they can describe what they want to do, not what they have done before and then get to do it at a cushy startup?
We're challenging the 'passive' vs 'active' paradigm. Most of the time we do exist in these binary states, rather somewhere along a spectrum between those two points. We think that if the opportunity is a good enough match - based on what you say you want - then you might be open to talk.
With regards to spam, no employer can message you with a notification trigger unless you make a clear choice to declare interest, after the match. So as a user you exist in two states - matched vs interested match. Only when you are an 'interested match' is your identity revealed to employers and conversation unlocked.
It drives enough traffic to kill a general web app. If you've just launched and have something like Rails / Django running on a single app server, it really doesn't that much to overload it. Wordpress seems to not handle it well either (unless you're caching).
I experienced a lot of lags and delays between pages with no feedback as to whether or not the page is loading. The tokenized skills input list had empty tokens when I clicked Confirm. The UI looks and feels very good in my opinion but the UX is broken.
Did you just split all engineering jobs in front-end and back-end? I'm afraid I never did a lot of neither, and I have no place to put my experience in there.
Would be nice if they listed how many jobs are in their DB. Last time I tried I couldn't login due to a server error. Next time I tried they didn't list anything for me. Even when I went back and added a little of everything instead of just coding. So as a user, I would never go back because it is several screens of work for no listings back. Maybe they should move to an email model where they will email you something if they ever get some employers on board.
This is the underlying model. We hope to find you matches on the spot, but if we can't then whenever a match appears we will notify you immediately via email that a match has been generated.
The majority of the jobs on the site are based in London right now, with some in New York and San Francisco. If you enter one of those locations you are more likely get a match.
On our end we are working to improve the way location is recorded so that we can cater for people with varying levels of specificity when it comes to where they want to work e.g. Anywhere, Europe, France, Paris
44 comments
[ 3.1 ms ] story [ 103 ms ] threadhttps://news.ycombinator.com/item?id=8652042
Worse, I do actually have a Github account, but I'm not going to tell you about it in order to log in and learn what this thing does, given that I have no idea what you plan to do with it. I can imagine a scenario where I'll discover a page on the internet proudly proclaiming that "Jason Is Looking For A Job!" when the reality is simply that "Jason is curious what dimensions go into that shape thing".
Give me a user/pass. Or better still, no requirement at all to dig in further. Then I'll dig.
I have yet to see a senior level .vimrc architect job in the wild.
Is that really true? I work for a mega-corp and I still put as much work as I can on GitHub. You always have side projects you can put up there as well.
That said, 99% of my contributions are in private repos, so my profile won't tell you a lot.
Nah, just keep on looking for work at a place that is willing to invest time in evaluating you.
I'd presume most people who come to HN are interested in evolving their skill set. The common worn-out office drone doesn't visit HN.
I have always used git, privately (eg: Bitbucket). And if I want to run an open source project, I can very well do that with Bitbucket too. I don't know why employers ask to see a Github profile.
I meant that they don't even know Git, let alone GitHub.
Most developers know about Github, of course, and have used it. You know, to get stuff.
What is the actual ratio of developers on github to not on github? All the developers where I work are on github, as well as the developers who I know personally. And if not on github, then bitbucket.
How many other people are the same?
Funny thing is that even if you create side-projects it's unlikely that you'll actually want to make them public. For example, say you want to create a Saas or bootstrap a small startup. The idea of putting all of your execution online for all to see (or borrow) is super scary.
For me it's about potential upside. If I take time off work to work on my own projects and I expend 10 hours a day in creating them, am I doing this in order to lock down a better job in a couple of months, or am I doing this because I would like to some day have personally crafted something which creates value for others and allows me to live freely?
I think you'd have to be crazy to want to put huge projects online for all to see (particularly as they can often require stigmatising relaxation of processes which don't make as much sense as an individual in order to be able to handle the workload as a sole engineer.)
Of course there are those that like to build development tools and libraries for other engineers. That's a priesthood I've not really entered properly - it's hard to create something which isn't just yet another me-too framework, but very respectable if you are able to do this (top ~1%).
NB: lot of MS guys on Stack - maybe a correlation there
We thought long and hard about what type of marketplace we wanted to create. We ultimately felt that there was a requirement to gate the marketplace on both sides to prevent bad actors from parasitizing the site (see LinkedIn). Engineers are in such high demand, that they go, inevitably recruitment agents and others will follow.
Github isn't perfect, but we our use of it is primarily as a signal of the type of people we want to attract (developers only) and as an additional admin hurdle for bad actors to have to go through.
So, yeah, we took a risk. Higher friction at the both ends (employers are curated, again to stop agents), but our aim is to create a high signal, zero noise matching service for engineers and this was the way to do it.
Our plan is to add further login options - bitbucket, stack et al. v2.
Hope that explains some thinking about the sign up flow - feel free to ping me any questions!
Hung
Shouldn't the content/workflow/UX of your site be what attracts certain kinds of users and disincentivizes others?
You make a very good point about content! Its true, that the content on the platform will determine whether the users get value from it. The caveat missing from your model is there are different types of users, with different, often opposing agendas, who might impact that UX.
Suppose for instance, we have a website where the content is generally great tech opportunities with cool companies.
Who's interested in that?
Engineers interested in job discovery might be and that is precisely our target audience. However, there are others who are also interested in that sort of information - recruiters, for instance, looking for leads. Now this is not the community we are trying to serve. In fact, we know that this demographic has the potential to seriously damage the UX of the employers of the others side of the platform (imagine if you signed up as a hiring manager to Workshape and it was advertised as direct employers only - and all you got in return was recruiters hassling you to give them a mandate to recruit for you...). So, we put into place mechanisms which firstly signal who our audience is, and secondly create friction to get in.
We understand that this is at some cost to rapid scale. But we don't value a platform which scales quickly, then equally quickly becomes toxic with the effluent of people who shouldn't be there.
Hope that makes sense! Keep the questions coming - feedback is welcome and necessary.
Hung
We know Github isn't a secure gate, but we also know that it is at least a gate - an extra admin layer to go through.
The analogy I'd use is similar to street crime. Your bike isn't secure because you put a D-lock on it. You're just making a little bit harder for the criminal to get it. A motivated bike thief? Yeah, he'll get your bike! But your casual criminal might just pass on rather than invest the time.
What we do want to avoid is to become a LinkedIn-like open platform, where recruiter parasitization is endemic. Our plan for a high signal, low noise platform is absolutely dependent on being agency free
Hope that answers your question. Keep them coming - all thoughts welcome!
Hung
Upfront an engineer might specify that they'd like to spend 60% of their time on Data Science and 20% in Architecture but this doesn't mean that they are skilled appropriately for this.
What about companies that want to hire people to fill roles that require very specific experiences and education?
Is this app assuming that we when asked to describe what we want to do, we won't describe anything too far from what we already do?
Is the market on Engineers sides so much that they can describe what they want to do, not what they have done before and then get to do it at a cushy startup?
the employer can reconcile the difference.
We're challenging the 'passive' vs 'active' paradigm. Most of the time we do exist in these binary states, rather somewhere along a spectrum between those two points. We think that if the opportunity is a good enough match - based on what you say you want - then you might be open to talk.
With regards to spam, no employer can message you with a notification trigger unless you make a clear choice to declare interest, after the match. So as a user you exist in two states - matched vs interested match. Only when you are an 'interested match' is your identity revealed to employers and conversation unlocked.
Ping me if you have any questions.
Best wishes
Hung
This is the underlying model. We hope to find you matches on the spot, but if we can't then whenever a match appears we will notify you immediately via email that a match has been generated.
The majority of the jobs on the site are based in London right now, with some in New York and San Francisco. If you enter one of those locations you are more likely get a match.
On our end we are working to improve the way location is recorded so that we can cater for people with varying levels of specificity when it comes to where they want to work e.g. Anywhere, Europe, France, Paris