HN Explorer

Ask HN: Who needs help with side projects?

37 points by throwaway83242 ↗ HN
In the spirit of other Ask HN : Hiring and who wants to be hired, I thought I would help people with their side projects. For free or otherwise.

Please post what your side project does and whether you need help with technology(frontend, backend, databases) or with marketing/growth hacking/sales etc.

Personally, I can help with Python, Java and other backend technologies.

64 comments

[ 2.7 ms ] story [ 133 ms ] thread
sharemywin
Interests:

Web3, OpenCV, NLP, using AI libraries and Apis, semantic search.

Languages: C#, Asp.net, Databases, some python, Played around with Django,

I was working on a way to scrape problems out of social media.

Also an AI idea generator.

hacky_engineer
I was working on a NLP(ish) idea for a while and got a descent prototype put together, but haven't had time to take the project past that. The idea was to train a model to classify tone of voice (a kind voice, angry voice, whiny voice, etc). Writing this out again now. I wonder if we could train a model to predict whether a statement is sarcastic or not..
sharemywin
The dataset would be the hard part.

I played around with Sentence Embeddings and Sentence Similarity.

I thought about being able to compare a sentence with "Happy" and get a score between them. This could be a generalized solution for a nlp problems.

I noticed it tended to match things a little different than I would expect. I would expect the subject of the sentence to match first.

hacky_engineer
I am working on a baby book called Computer Engineering for Babies (https://computerengineeringforbabies.com). The biggest question I have is an SEO/branding one. I have about 3 brand names, and I don't know how to structure them or which ones to get rid of, and how to have a cohesive brand. Anyway, I'm looking for someone that can coach me on some SEO/branding stuff..
kleer001
that doesn't sound like it's within the scope of OP's request, but best of luck!
digisign
Looks fun, how can I get a couple without having to wait until December? Seems like that would be an easy way to boost conversion, to have them come in once a month. Guessing you don't have enough volume?
bhu1st
I am working on weekends occassionally on a side project of mine, a real-time version of board game Baghchal (http://baghchal.net). The stack is Node.js/MongoDB.

I have identified couple issues that I'd love to work on at https://github.com/bhu1st/baghchal.net/issues but I lack time/resources.

Many thanks in advance for any technical help/support/collaboration. Please DM me.

meesterdude
My project is a new way to browse the web, spatially. Built in unity/c#. Mostly done but there's always more to do so open to help with hte dev, Would also be great to work with a Ruby on Rails dev on building out the platform side.

would also love help building up traction/excitement with users/potential customers.

email is violentpurr at gmail if you're interested. I'm based on the east coast.

ternaryJimbo
Interested! By spatially do you mean this is a vr app?
meesterdude
Sort of! It's in 3D. Spatial is leveraging the human ability to reason about & remember landmarks as part of navigation of a space and associate information.
Optimal_Persona
The advent of JavaScript endless scrolling was really disorienting for me, it made me realize I had an almost "Palace of Memory" approach to mentally categorizing online information by its navigation and page structure before then.
pm2222
There are columns of data, and as a simple example they are {{1,a},{1,b},{1,c},{2,a},{2,b},{2,c},{3,a}} of two columns and seven lines. I'd like to process the data through a function which would yield {{{1,2},{a,b,c}},{{3},{a}}} which is slightly better than {{{1,2,3},{a}},{{1,2},{b,c}}} because former has 7 leaf elements vs latter 8 leaf elements while both has if you will two lines.

How to write that function?

lesuorac
I think you need to provide more information.

Do you have a CSV file? Is this a specific data structure in some language?

Is the output another CSV file / std out?

petronio
If the data is ordered by row as in your example (e.g. {1,a},{1,b}, etc), a relatively straightforward option would be to read all columns for the row and use the values as a key such that hash(columns) -> list(rowNum).

This may not give you the lowest number of leaves depending on your dataset. If you're looking for the absolute smallest number of leaves, you could use each column as an individual key and start intersecting their lists to see if that produces a reduced leaf count.

minhmeoke
This is similar to a Karnaugh Map: https://www.youtube.com/watch?v=3vkMgTmieZI

Basically, you can map the data values into a boolean truth table representation:

      a b c
    1 x x x
    2 x x x
    3 x
And use the Karnaugh Map to visually simplify:

      a b c
    1 ┌───┐
    2 └───┘
    3 □
into:

    {1,2}{a,b,c}
    {3}{a}
If you'd like to do this algorithmically, there are Karnaugh solver libraries which use algorithms such as the https://en.wikipedia.org/wiki/Quine%E2%80%93McCluskey_algori...
pm2222
Very nice. Can this be applied to more than 4 dimensions?

This problem is an abstract of a network firewall rule optimization. The a,b,c... are for source IP/net, destination IP/net, protocol/port, permit/deny and probably more like zone names.

mannanj
Your Life in Blocks: A concept inspired by Tim urban's Your Life in Weeks and by my weekly poster which I write a few blocks of text in.

I wanted a way to track significant things that happened in my life, and looking back at the physical poster, it has quite the impact on my perspective but I noticed it's a hassle carrying that thing around.

Angular, Firebase, Sass, NGRX Store: The MVP was functional for this stack but only for one account and without Firestore rules written. It has since been defunct and I've been stuck on scaling it to work for more than one user without public database rules.

Site: https://yourlifeinblocks.firebaseapp.com/user

Git (open repo): https://github.com/mannanj/life-in-blocks

thenewwazoo
I've got an open-source Homebridge plugin (https://github.com/thenewwazoo/homebridge-lutron-caseta-leap) that has a bug in its client-handling code. There is some problem with connection handling that is causing user pain. I am an experienced developer, but new to Typescript, and absolutely cannot chase it down. I would love the help of someone with Node socket coding knowledge willing to take a look.
jka
I've been developing a recipe search engine and meal planner at https://www.reciperadar.com/ -- all the code is free and open source.

I think the two largest challenges it has at the moment are:

- Developer experience / infrastructure setup: the service is Kubernetes-based, composed primarily of Python+Flask microservices, and although the setup steps are documented[1], they're not yet automated.

- User interface: my lack of user interface design skills are fairly apparent, I think. Having some clear UI goals to work towards would be helpful - there's likely some overlap with product design questions here (what should be the range of functionality that the the app offers?)

To be transparent: although I hope to offer compensation in future, the project is zero-revenue currently, and I don't feel ready to provide any monetary returns yet. If that hasn't put you off: any feedback is welcome, from discussion comments to pull requests to business model questions/critiques!

PS: Don't forget to mention one or two of your own side projects if there's anything you're looking for help with, too.

[1] - https://github.com/openculinary/infrastructure/

PhillyG
I wouldn't say your UI is bad - it's certainly a reasonable MVP at this point. You could probably stick ads on it or a newsletter sign up on it for people interested in it. If I wasn't looking for UI improvements to point out, first glance would be it's pretty good, and what I'd expect - I certainly wouldn't immediately bounce - though the input boxes do need to support splitting ingredients by typing a comma - typing enter is not instinctive and even once I'd figured that out, when going again I "messed up". I listed ingredients as I'd expect to for an input box, and when I moved to the next box my input was lost.

It sounds like you need to spend time using it yourself, thinking over it's use, getting others to use it, watching them get stuck, listening to their questions. Get family or friends who have never used the app to use it while you are there and find out what they struggle with. For knowing what functionality it needs - find people who are interested in paying for that kind of product - family and friends are great for doing the app equivalent of proofreading, but unless they are in your market they may well have lots of ideas, but you'll have no idea if they're the right ideas to build.

Some confusing things about your UI (in order of me noticing them):

- "Are there any ingredients that are not available?" - this looks like it needs rewording to something like "avoid these ingredients" or "any common ingredients not available", I knew what you meant, the search will exclude recipes with those items, but it felt weird to read - there are probably 1000s of ingredients that I don't have in my kitchen, I'm not going list those. I might list things like milk if I where cooking for a lactose intolerant person in that box, or I might use that for if I where out of something common.

- on laptop screen, box around the search form has a huge gap to the right of the form inputs. I'd either make the inputs a % width of that box, or put them in two columns when wide enough

- the previously mentioned thing about inputs needing to treat , being entered as if the user had tapped space

- I'd probably expect to see filters for vegan/veggie/pescatarian / low gi etc but then maybe not - that's the kind of thing you could build if you run out of things from proper users - unfortunately I rarely cook proper meals so I'm not really in your target market I would assume.

- Shopping list feature is a feature I could use, if it where better, (though I currently use a shopping list app that I threw together myself by quickly modifying another thing I'd made previously, and it works reasonably well for me) - your icons seem too small and fiddly, and I'd want a few buttons on screen of common things - so I can tap those instead of typing them.

jka
Thank you for the response!

> support splitting ingredients by typing a comma

That sounds sensible, yep; filed as https://github.com/openculinary/frontend/issues/210

> "Are there any ingredients that are not available?" ... I knew what you meant, the search will exclude recipes with those items, but it felt weird to read

That makes sense too. If I remember correctly, that prompt was most-recently rephrased during a pandemic-related lockdown (with subsequent unpredictable ingredient shortages), and so that context may have affected the choice of wording; either way, I agree that it's odd phrasing and should be updated.

Hopefully that'll be a relatively quick correction, although it will require internationalization (currently machine-translated without review by native language speakers, not ideal); it's filed as https://github.com/openculinary/internationalization/issues/...

> box around the search form has a huge gap to the right of the form inputs

> I'd probably expect to see filters for vegan/veggie/pescatarian / low gi etc but then maybe not

Two good points here, and possibly combinable. Perhaps those dietary recipe filters could be placed in the excess space available next to the search controls.

Today the search API does theoretically support filtering[1] on a few dietary properties -- but that functionality isn't yet visible and available to the user.

Feature tracker filed as https://github.com/openculinary/frontend/issues/211

- Shopping list feature ... your icons seem too small and fiddly, and I'd want a few buttons on screen of common things - so I can tap those instead of typing

The frequently-added-items shortcut idea is smart. The shopping list feature and the meal planner are under-attended components of the app relative to the recipe search/explore components, in my opinion. Let me think about this for a while, there are a few considerations and I'd like to be concise when responding about this in more detail.

[1] - https://github.com/openculinary/api/blob/72075f66cd6fda5b809...

jka
Hey - if you re-read recent threads and see this reply, great; and no worries if not - thanks for your feedback, in either case.

All I'll say for now is that arguably the shopping list should become an important nexus within the app. Like an accounting entry-book with both order history and also yet-to-be-filled orders, it could be used to approximate the kitchen inventory at points-in-time. That should become powerful when changes in inventory are used to feed back into current and future meal planning.

Separately: advice noted about testing with friends and family. My partner and I use the app at home once or twice a week, and following recent improvements it's probably time to reach out for more feedback from others.

I find maintaining a balance between asking for (re)evaluation of the app and a hope for regular everyday conversations a little tricky; but to be honest I've always found the latter challenging, probably.

shakezula
In a similar vein; if anyone is looking for an accountability buddy on their project, hit me up!
jawmes8
What's your contact?
shakezula
Just updated my bio with a good email to reach me at.
jms55
I'm working on a Screeps-like game using WebAssembly. You compile a script to WebAssembly, and for each robot you control on the server, it runs the script in an isolated environment to choose an action for that robot. The goal is to write a program to coordinate your robots into gathering resources, replicating, and expanding your control of the server.

Here's an example bot script: https://github.com/JMS55/botnet/blob/master/example_bot/src/...

The basic infrastructure of the project is more or less in place, besides a visual replay viewer which I'm working on right now. What's needed is a bunch of work in designing game mechanics and APIs. I don't actually have any plans at the moment beyond bots running around and harvesting randomly generated resources. Feel free to open a discussion on the github page if you're interested in Rust, WebAssembly, and video games.

https://github.com/JMS55/botnet

Rodeoclash
This is cool! I've had this idea knocking around in the back of my head similar to this about an MMO like game where users could write WASM code and "imbue" items in the game with it. Sort of allowing them to create interesting game logic to build stories and things in it.

Elixir always felt like it'd be a good backend to coordinate running things

jms55
Dang, that's also a neat idea! I do also like the Elixir language. I'll have to keep this in mind for future projects. Perhaps I can integrate some sort of system into botnet, where you can program custom bot actions somehow.

It would make the replay system I'm writing much more difficult though xD

mikercampbell
Upvote on Elixir! I'm learning it right now at Divvy and loving it!
Rodeoclash
I have a video player designed for esports coaches to do gameplay reviews of matches that their team has played. It has the unique feature that you can synchronise multiple viewpoints from a match together. It's free and open source.

https://github.com/Rodeoclash/vodon-pro

I've built it using Electron as an MVP but I'd like to rebuild the most successful features into a native app using gstreamer as a V2 version at some point.

I'm looking for help with reaching more esports teams that might be interested in using it, graphic design feedback (I've done it myself :/ ) and someone who has knowledge building native apps that I could bounce ideas off.

muzani
http://random-character.com

It generates characters from tropes, weighted by the popularity of those tropes. Which was a nice experiment, but it's probably not the future direction.

This is something with a lot of iteration. Procedurally generated stories and backgrounds. Letting users input a character and using the tool to shape the character around it. Relationship graphs to other characters.

AI will be in use, but it functions more as a renderer, with good old algorithms forming the structure. But this needs a lot of development work and involves taking money from users, at the very least to pass the cost.

Looking for: Anyone interested in writing or AI content generation. Will profit share but currently profit is $0.

Traction: Approx 2000 users/month, mostly from Bing. No marketing besides mentioning it on a forum twice a year.

aviditas
I really like this idea! I feel like you could have the character mapping and input be geared towards writers for a story documentation tool as a UI and framework for the generated content would be in what a writer would need. Then offer as much or as little generator access as the user wants.

For that generated content, I do a lot of character creation in a completely different context than writing/D&D, and I would pay for a service or software that let me input or override the story type character data with ones of my own making.

The goal of self-sustaining stories that can be tweaked or woven into is awesome.

muzani
Oh you popped on the Discord yesterday. Sorry about that, Discord is a poor replacement for the old anon chat rooms, lol.

Thanks for the feedback. I might just pivot straight into that. I was dragging my feet on it because I didn't get much people interested in that, but only because my only communication medium is the survey.

aviditas
No worries at all. Speaking of Discord, there are a ton of writing and D&D communities that would be great resources for feedback. You have an advantage in that area as you aren't charging for your current product and so hopping in to solicit feedback won't get you the same kind of backlash as long as it's okay in the groups' rules. Feel free to message me; I wish you the best of luck!
josephrmartinez
I'm building a simple web app to help me spend more time actively creating ideas when I'm online and less time just passively consuming information.

The app is super simple. It just encourages you to write ten ideas on a topic. You can either come up with the topic or borrow a pre-existing topic from someone else on the site. When you finish coming up with ten ideas on that topic, you can see all the other ideas that everyone else has come up with on that topic.

So instead of just scrolling and never contributing anything actively, I have to at least think about the topic and work on it a bit if I want to see what other people are thinking as well.

I can select a topic prompt "10 ideas for fun weekend DIY projects that cost less than $300." I write my ten ideas and then I see that 6 other people have also written ideas. Bam! Now I can see 60 more ideas for weekend DIY projects.

I've put together a very rough mock up using Anvil here: https://ten-ideas.anvil.app/ (Anvil lets you build web apps using nothing but python).

I'd love some feedback on what to do next for: - finishing the project on Anvil - finding about ten initial users to give feedback - sensible ways to put this out in the world

I just started learning to code a few months ago. My profession is farming, this is just a fun, non-commercial side project to learn and encourage positive online habits. Any pointers would be massively appreciated.

chitowneats
Bravo Joseph on learning to code and using your new skills to build something useful. If you want some feedback I'd be happy to provide it. Email real10xdev at g mail dot com and we can do a review of the project (and the code, if you want) over email or voice chat. Whichever you prefer. I have a decade of experience in the industry building web applications including with python.
frozencell
A model that gives hackernews feed a two-layer ontologic classification of articles. The first layer is the type of medium (ie. ebook, blog post, research paper, question, ...) and the second is the category content itself (ie. health, programming, ...)
throwawaygog6
Sounds interesting. I am working for something similar. Can you give some way to contact you.
samsquire
I wrote a dynamodb querying, SQL and Cypher database in Python. The code is very small and simple HTTPS://GitHub.com/samsquire/hash-db

It supports distributed joins over SQL. I plan to add multimodel specifically GraphQL and document storage.

I also wrote a multithreaded user space M:N scheduler in Rust, C and Java. It multiplexes N lightweight threads over M kernel threads like Golang goroutines. I also wrote an evented version with a multiconsumer multiproducer RingBuffer that handles IO. I need to add epoll network events and you have an application server. This is like Libev or libuv an event loop for processing and queuing up blocking syscalls without blocking.

I want to implement a LLVM JIT runtime. I have an idea for a very simple event queue that handles multithreading while being threadsafe. One of my other projects is multiversion concurrency control.

If you find this comment interesting you might enjoy my blog. https://GitHub.com/samsquire/ideas4 https://GitHub.com/samsquire/ideas3 https://GitHub.com/samsquire/ideas2 https://GitHub.com/samsquire/ideas

pbronez
That’s super cool! Earlier this week I was getting frustrated with NetworkX and started wondering: “what’s the easiest way I run proper graph queries on this in-memory dataset?” This seems like a reasonable option, will add it to my list to try out.
samsquire
I warn you the code isn't ready for serious usage. It's more of a proof of prototype to show that these technical problems can be solved very simply.

So I doubt it would handle a challenging query! It only supports the bare essentials.

is_true
Which is the cheapest way to serve Json to customers with a budget under usd10 a month, a Json size of aprox 50KB and aprox 1 million reqs/day
minhmeoke
Netlify free tier + Cloudflare CDN (if requests are cacheable).

https://www.netlify.com/pricing/

https://www.cloudflare.com/plans

is_true
Thanks. Something I missed is that the JSON is gonna be updated every 1 minute.

I've never tried netlify or vercel, I've been told that the 1 min updates are a problem in platforms like that. I'm an nginx kind of guy but I'm helping an org that has no budget whatsoever and I'm exploring options.

minhmeoke
Yes, netlify typically does deployment by pushing to a git repo or uploading a zip, the update latency could take several seconds to a few minutes.

Another option is to use something like Firebase:

https://firebase.google.com/docs/hosting

The free tier gives you 2M cloud functions per month, then $0.40 for each additional million cloud function invocations:

https://firebase.google.com/pricing

Update latency is a few hundred milliseconds which should be sufficient if you are sending updates every minute:

https://stackoverflow.com/questions/37930219/is-firebases-la...

Just be aware that if your JSON payload is large, then you might consume all the egress bandwidth and it might end up costing more than you expect.

There are also alternative platforms which provide similar functionality (I have not used either of these though, so YMMV):

https://backendless.com/pricing/backendless-cloud/#compariso...

https://azure.microsoft.com/en-us/pricing/details/functions/...

throwawaygog6
That's ten requests per second, with some caching I am assuming. How abou a digital ocean droplet for $5/month?
is_true
That's what I proposed but there's a new designer that wants to use shiny tech. I'm just an advisor for a nonprofit, they ask me because I do it for free basically.
colthn
I'm working on an in-browser Doom-like level editor. I'd really appreciate some constructive criticism of the architecture or anything else that jumps out for rewriting besides some general code cleanup.

https://bitbucket.org/bbcolt/unnamed-project2/src/master/

jka
Perhaps less of a code-related comment than you are looking for, but: to make it easier to collect feedback on the architecture/code, I think it might help to add a readme document that summarizes the layout and structure of the repository in a few paragraphs.

Yes, there's a bit of overhead to maintaining that, and it can fall out-of-sync with the code over time.. the key benefit, though, is that it provides useful context for reviewers, a bit like a geographic map does when navigating a new place.

The goal should be to make people feel like they understand more of the purpose and design of the project shortly after they arrive; as a result they'll be able to figure out what aspects of their own experience they can use to help review (and perhaps contribute changes to) the code, and where to go (in terms of files/directories/modules) to best apply that expertise.

xieweizhi007
Hi, my friend and I have a side project that let users learn almost anything for free online. We are both front-end developers and want a backend developer to join. If you are interested and want to learn more. Just drop me a message deexie1990 # gmail.com
zzo38computer
I am working on "Free Hero Mesh" project (but it is not written in Python or Java). It is a puzzle game engine, and is FOSS.

Things that I can use help with are including (but not necessarily limited to):

- Promotion of project

- Testing

- Bug reports and feature requests

- If you are interested to make something with this

- Porting and distributions

- Patches to the code, if necessary

- Maybe someone has idea to improve documentation (such as tutorials) or to make a logo, or something else

http://chiselapp.com/user/zzo38/repository/freeheromesh

zzo38computer
Currently, I am the only person on this project. If others might also join with this project, then this might help, too.
melezhik
http://sparrowhub.io:2222 - SparkyCI - dead simple CI service.

Please give me a shout if you need to add other languages support. I would consider such an interest as a help )))

So far the service supports amazing Raku language, but the platform is general enough to support other languages as well …

Also I would appreciate help with frontend … The UI is simple html plus some JS / css under bulma.

pdyc
I am working on creating a hosting server that can cater to landing page+blog+selling digital content with minimal maintenance overhead.

Beta version is ready here https://github.com/newbeelearn/sserver

I am looking for following - marketing and sales (I need some guidance from experienced marketer who has done this before) - beta users to try this product and give feedback so that i can improve - frontend for admin UI (backend api's are ready)

I can help with golang based backends if that is needed by anyone.