Ask HN: Anyone joined a company after contributing to their OSS projects?
38 comments
Yup.
Looking to develop my Rust skills I found a OSS project to contribute to. Learned a ton working on an issue. A few months later I found out the company was looking to hire so I applied on the basis of my contributions. There was no interview, the company just took me on straight away. I'm still there and loving it!
Looking to develop my Rust skills I found a OSS project to contribute to. Learned a ton working on an issue. A few months later I found out the company was looking to hire so I applied on the basis of my contributions. There was no interview, the company just took me on straight away. I'm still there and loving it!
At PostHog we've hired a bunch of people like this.
One word of caution - sometimes we get a contribution from people right ahead of applying for a job. The best applicants are ones who are existing users and who really are interested in the product, versus those who are doing a pull request to stand out. The latter is still a positive, but the former (depending on the quality of PRs!) is the home run!
One word of caution - sometimes we get a contribution from people right ahead of applying for a job. The best applicants are ones who are existing users and who really are interested in the product, versus those who are doing a pull request to stand out. The latter is still a positive, but the former (depending on the quality of PRs!) is the home run!
I personally know a few at Mozilla who used to be regular contributors to Firefox.
And I know a somebody else who used to contribute to the Linux kernel before they eventually ended up at RedHat.
I don't know if it's because of the contribution that they got hired, but it could have been a factor.
-----------------------
Also there's that tangentially related story about Homebrew creator not getting a job at Google, so there's that. [1]
[1] https://news.ycombinator.com/item?id=15981338
And I know a somebody else who used to contribute to the Linux kernel before they eventually ended up at RedHat.
I don't know if it's because of the contribution that they got hired, but it could have been a factor.
-----------------------
Also there's that tangentially related story about Homebrew creator not getting a job at Google, so there's that. [1]
[1] https://news.ycombinator.com/item?id=15981338
That's pretty much how we hire at Stability AI.
Contribute to LAION, Eleuther or any of the image/media generation open source notebooks and you'll get an interview pretty quick, hired dozens that way.
Contribute to LAION, Eleuther or any of the image/media generation open source notebooks and you'll get an interview pretty quick, hired dozens that way.
Emad, what are next steps for Stability AI? I’m into the Data Science education space (exited my prev company in ‘19). If you’re looking into something edu-related, let’s connect.
I contributed to mediawiki, prior to being hired by the Wikimedia foundation (this was a while ago, i dont work there any more, although i still do open source contribs sometimes). I definitely wasn't the only one who contributed before being hired.
There are downsides though - if you're really into the open source project, and then get hired it starts to be less fun. It also means that if your hired to work in one specific area, you probably stop doing other things because you need to get your team goals done, and it feels silly to do other stuff, which can be kind of sad if previously you were doing a variety of stuff.
There are downsides though - if you're really into the open source project, and then get hired it starts to be less fun. It also means that if your hired to work in one specific area, you probably stop doing other things because you need to get your team goals done, and it feels silly to do other stuff, which can be kind of sad if previously you were doing a variety of stuff.
Commercialized FOSS projects with a pool of contributors have a talent pool to choose from when hiring that non-free projects can only dream of. How else can you hire someone who (1) is already familiar with your code, (2) already knows your team, and (3) has already been proven to be good at the role?
SourceHut doesn't often hire full-time but we do often find ways to get funding for people on a part-time basis, and 100% of the time we already knew who those people were going to be because of their work in the community. It just makes sense.
SourceHut doesn't often hire full-time but we do often find ways to get funding for people on a part-time basis, and 100% of the time we already knew who those people were going to be because of their work in the community. It just makes sense.
I’ve contributed to the Supabase JavaScript client libraries and built some starter templates before joining the team https://twitter.com/thorwebdev/status/1569630009240993793
Back when I was monitoring deno's progress, I came across this PR by Aaron O'Mullan that improved deno's perf significantly: https://github.com/denoland/deno/commit/3d2e05dc7b6735fe0b81...
I believe they joined deno shortly after.
So, I guess it does happen.
I believe they joined deno shortly after.
So, I guess it does happen.
Out of curiosity I went through that whole pr discussion and I guess a maintainer posted a performance comparison, but Im not sure what numbers I'm supposed to be comparing. Do you know?
https://github.com/denoland/deno/pull/9722#pullrequestreview...
https://github.com/denoland/deno/pull/9722#pullrequestreview...
I think it depends a lot on the company. If their oss project doesn't get much love and someone is trying to force their way in, I don't think it will work out well.
But Posthog for example have oss in their DNA and I was using their product, communicating with their excellent support and fixed a tiny bug in their Ruby SDK. A while later they reached out with a job opportunity. It didn't work out in the end but it was a fantastic experience!
But Posthog for example have oss in their DNA and I was using their product, communicating with their excellent support and fixed a tiny bug in their Ruby SDK. A while later they reached out with a job opportunity. It didn't work out in the end but it was a fantastic experience!
Speaking from the company side, this is our main method of hiring! It works great.
A key factor would be that we as a company support and maintain an open source framework. I'm not sure how well it would work without that.
A key factor would be that we as a company support and maintain an open source framework. I'm not sure how well it would work without that.
I know a person who got a great paying job at a streaming service company, by working on a set of open source tools heavily used by company. He didn't have to do any sort of technical interview. Now he works on those open source tools full time remotely, and got about a 70% pay increase.
I made https://www.jooq.org. Then hired myself in the company I created to maintain jOOQ
Minor contributions I made to Rails in 2007 led to being recruited as the first tech lead and eventual CTO for a nascent startup. There was no premeditation here though, it was just serendipity. I've found most big breaks in my life this flavor. It's more about being open to opportunity so you can recognize it when it comes rather than trying too hard to orchestrate it.
I wrote SupaScript[0] before joining Supabase.
[0] - https://github.com/burggraf/SupaScript
[0] - https://github.com/burggraf/SupaScript
I've gotten an offer from OTRS AG after contributing some to their core (back then Open Source) ticket system.
I've declined, and a few years later they basically closed-sourced their platform (it's still source available, but only with years delay, and they no longer accept feature contributions). Can't say I regret my choice.
I've declined, and a few years later they basically closed-sourced their platform (it's still source available, but only with years delay, and they no longer accept feature contributions). Can't say I regret my choice.
Contributed to the Supabase Python SDK[0] and postgrest-ex[1] before joining early Jan 2022
[0]https://github.com/supabase-community/supabase-py [1]https://github.com/supabase-community/postgrest-ex/commits/m...
[0]https://github.com/supabase-community/supabase-py [1]https://github.com/supabase-community/postgrest-ex/commits/m...
Most of the developer at my company (Nextcloud) are former contributors to Nextcloud or other open source projects. So definitely yes.
I hired my first contractor through his contributions on our product.
We talked and hit it off instantly. He's very passionate about the product and when he mentioned he is looking for work, I stepped up and hired him.
I'm so very happy about getting him onboard.
We talked and hit it off instantly. He's very passionate about the product and when he mentioned he is looking for work, I stepped up and hired him.
I'm so very happy about getting him onboard.
Indirectly.
I worked on a project not owned by a single company. But Red Hat hired me to work on it after, I worked on it at another company. :)
Alas, since I haven't found a good OSS gig. But that's life.
I worked on a project not owned by a single company. But Red Hat hired me to work on it after, I worked on it at another company. :)
Alas, since I haven't found a good OSS gig. But that's life.
I made contributions to Supabase before joining the company, although my contributions weren't to any of their GitHub codebases but more of a community one where I answered questions on their GitHub discussion board on a regular basis for users who were less experienced with the platform than I was.
A few former coworkers of mine were improving OSS code for their employer as user/contributors then recruited to the maintaining companies to work on that code full time.
definitely. i used to run the visual studio / vs code teams for python, R, nodejs, jupyter etc at microsoft. working closely with and hiring from the os community was both delightful and a priority for me. several key employees were hired that way.
No to the title question, but yes this definitely happens - someone from Cloudflare emailed while I was contributing in quite early days of (that's not a brag - I mean it was relatively easy for me to make quite large/impactful/sweeping changes to) their Terraform provider.
(And I perhaps wish I'd gone for it! Was poor timing as I'd only just started elsewhere at the time.)
(And I perhaps wish I'd gone for it! Was poor timing as I'd only just started elsewhere at the time.)
Yes, it's how I got my job at Red Hat, after contributing in the Hibernate space for a while. Same for many other Hatters.
Companies have never offered me a job based on the projects I've contributed to but I haven't contributed to many projects to begin with. People have reached out to me because they saw my own projects on GitHub though. I wasn't able to accept a position at the time but it was really nice to be recognized.
We have hired quite a few people on this basis at Docker. Always on the lookout for more.
Long long long time ago I joined Sun after contributing to GlassFish and NetBeans for a while (documents, test cases, some new plugins for NetBeans). They approached me and hired me to work on GlassFish application server.
Yup.
At WunderGraph, This is exactly how we found one of our technical co-founders and how we actually got our first hires.
We're huge advocates for Open Source and actively look at users who are contributing to our repo as potential hires.
At WunderGraph, This is exactly how we found one of our technical co-founders and how we actually got our first hires.
We're huge advocates for Open Source and actively look at users who are contributing to our repo as potential hires.
Been working at Hasura for almost 3 years now, this was how I joined.
I contributed to Supabase Realtime [0] before being hired late January of 2021.
[0] https://github.com/supabase/realtime/commits?after=9990120c6...
[0] https://github.com/supabase/realtime/commits?after=9990120c6...
I helped Supabase create and maintain their Flutter/Dart library before joining them mid April of 2022! https://github.com/supabase-community/supabase-flutter
I'm in the reverse situation, my anacrolix/torrent, DHT and other projects are used heavily and I work on and with them there.
if the org has funding and you made a thoughtful contribution, I can’t imagine a better signal for knowing you’d be a good hire
Yeah, that is how I accelerated my career ... was back in 2006 though. Still think this is a great path to follow though.
Yup, got interviewed by person who reviewed my feature PR on their OSS. Easiest interview ever.
My current and last company both did this. This is definitely a thing.
I think Andreas Kling (of Serenity OS fame) did this
If you already work for them for free, why should they pay you?
This dumb question has an important answer: accountability and longevity.
It's the same reason CIOs of large companies pay more to work with companies that provide a Service Level Agreement: their bosses (aka the board of directors) wants to know that there's someone smart who will pick up the phone if there's a real problem, 24/7. And someone to sue if things go south.
If someone is making excellent OSS contributions to libraries that you rely on, it is legitimately unsane to not try hard to hire them. It means that they won't be recruited by someone smarter, distracting them from further contributions. Also, if you bring these folks into your culture, it's almost always a huge win/win because they are already up to speed without training. They are already true believers, and you're rewarding their early belief in your vision by unblocking them so that they have their needs covered, allowing them to be even more productive.
You'd be hard pressed to find a more potent way to hire top talent. The idea that you are somehow getting ahead by not hiring someone who wants to be hired is probably the most expensive wrong idea you hold.
It's the same reason CIOs of large companies pay more to work with companies that provide a Service Level Agreement: their bosses (aka the board of directors) wants to know that there's someone smart who will pick up the phone if there's a real problem, 24/7. And someone to sue if things go south.
If someone is making excellent OSS contributions to libraries that you rely on, it is legitimately unsane to not try hard to hire them. It means that they won't be recruited by someone smarter, distracting them from further contributions. Also, if you bring these folks into your culture, it's almost always a huge win/win because they are already up to speed without training. They are already true believers, and you're rewarding their early belief in your vision by unblocking them so that they have their needs covered, allowing them to be even more productive.
You'd be hard pressed to find a more potent way to hire top talent. The idea that you are somehow getting ahead by not hiring someone who wants to be hired is probably the most expensive wrong idea you hold.
code ownership could also be a factor. if a contributor gets hired, their new code gets owned by the company
Imagine you make $300 for every hour someone spends and you pay them $100. Crude, and you can disagree with the numbers, but this is the general idea when employing people.
You get 4 hours a week for free, you make $1200 - $0 = $1200
You get 40 hours a week for $4000, you make $12000 - $4000 = $8000
By hiring them, you make more profit.
You could try to get 10 people doing 4 hours a week for free, but maybe there is a limit to the number of people willing. Let alone the coordination and efficiency advantage of dedicating more time per week to tasks.
You get 4 hours a week for free, you make $1200 - $0 = $1200
You get 40 hours a week for $4000, you make $12000 - $4000 = $8000
By hiring them, you make more profit.
You could try to get 10 people doing 4 hours a week for free, but maybe there is a limit to the number of people willing. Let alone the coordination and efficiency advantage of dedicating more time per week to tasks.
Pay another person to work 40 hours and now you make $9200 instead of $8000. By hiring them you are losing the work you would have gotten for free. As you said there is a limit for how many people are willing to work for free so you shouldn't get rid of the ones willing to work for free.
You also get to align their projects within your codebase to the rest of your efforts and strategic vision by way of paying them.
You can just talk with them to figure out something high impact that they would like to work on.
You can try. Let me know how that works out for you, especially when you need someone to fix a small niggling but uninteresting bug.
This is an extremely shortsighted comment.
If they don't pay you they:
- Have no incentive to make you continue contributing,
- Have no decision power over what component you contribute to,
- Have no decision power over when or how often you contribute,
- Have no ground for asking that you follow any particular process (due to the first point),
- etc.
If they don't pay you they:
- Have no incentive to make you continue contributing,
- Have no decision power over what component you contribute to,
- Have no decision power over when or how often you contribute,
- Have no ground for asking that you follow any particular process (due to the first point),
- etc.
All of these can be accomplished via social engineering instead of paying.
They have to pay somebody to do the social engineering too.
Because they are under no obligation to continue contributing, and if the organization is serious about the future of the project and spot some talent, they'll want to try and get that person into a more permanent role. The real question is actually the opposite of the one you asked: why wouldn't they want to try and pay you for your work if it's beneficial to the product/company?
Plus you may not be working on exactly what they need from the project, but you already know the domain, the code base and the people. Even if they let you keep a few hours to do the stuff you were doing before, it's a net gain.
If your question is a sincere one, then you're asking, of employees who have been approached by companies via OS contributions, "why did they do that when they could have got that work for free?". Hearing from employers who have done this might be interesting, but I'm not sure that the employee would be the right person to ask.
On the face of it, however, to me at least it reads like your 'question' is less interesting: a rhetorical restatement of "companies don't do this". Then the 'evidence' you offer is an invented just-so story. Like all such stories it reflects your response to the world and the domain of your imagination, but tells us nothing about the world itself. It's a plain empirical fact that companies either do this, or they don't, and you find out which is the case by examining reality, not by weaving tales.
Telling stories to assert that the world 'must' be as we perceive it is common but without worthwhile uses outside of debate halls.
On the face of it, however, to me at least it reads like your 'question' is less interesting: a rhetorical restatement of "companies don't do this". Then the 'evidence' you offer is an invented just-so story. Like all such stories it reflects your response to the world and the domain of your imagination, but tells us nothing about the world itself. It's a plain empirical fact that companies either do this, or they don't, and you find out which is the case by examining reality, not by weaving tales.
Telling stories to assert that the world 'must' be as we perceive it is common but without worthwhile uses outside of debate halls.
I am not offering any evidence for my position, but I am saying that not hiring seems like it could be a more optimal business move.
Which is a very odd response to a question of fact. Rationalism 'ain't dead yet, it seems.
> If you already work for them for free, why should they pay you?
Because they'd like you to do it full-time instead of only in your spare time?
Because they'd like you to do it full-time instead of only in your spare time?
But why not hire someone else to do it full time so you have full time + part time.
Because the part-time volunteer contributor has already demonstrated aptitude and interest in the project, they don't even need to be interviewed or trained.
For large companies one of the obvious value-adds of OSS is it optimizes out the whole time-consuming, costly, and error-prone recruiting/hiring/training process. That all occurs voluntarily on the contributor's time who's so interested in the work they've already sought out the job, developed a rapport with the collaborators, and have been doing it for free.
For large companies one of the obvious value-adds of OSS is it optimizes out the whole time-consuming, costly, and error-prone recruiting/hiring/training process. That all occurs voluntarily on the contributor's time who's so interested in the work they've already sought out the job, developed a rapport with the collaborators, and have been doing it for free.
Because then you'll have to do whatever is the priority that they think you should be doing, instead of doing whatever you previously wanted to do with it.
I'm just curious whether companies are open for this. Lately there is more traction towards open-source and whether a person can start contributing to it and later join the core team.
one personal example i've read is the developer of nomad levant project was acquihired[1] by hashicorp
[0] - https://github.com/hashicorp/levant [1] - https://twitter.com/jrasell/status/1230132251893125120