I gave the AI arms and legs then it rejected me
grell.dev
grell.dev
Simply put, anything not a viral license like GPL allows parasitization by companies effectively living off FLOSS devs, with absolutely nothing to gain. Human rights under GPL were meant to apply to humans, not '3 lawyers in a trench coat' (corporations).
They can make their decisions (snubbing a dev of code they deem good enough for enterprise). And you can make comparable decisions, punishing them for the sheer hubris.
It also reaffirms that my decision of AGPL for everything is the right one. They can contact for custom terms.
Since your replies below are focusing on compensation: have you actually made a nontrivial amount of money with that model?
I would expect that should be a prerequisite to reaffirm it was the correct decision, especially if you're giving unsolicited advice to strangers about how they should license their software.
The author said he was proud of this outcome and nervous at how widely his hobby project will be deployed. That sounds like the ambition of many open source authors and a win. Might never have happened with GPL.
This has been debated a lot already. It depends whose freedom we are talking about. It is overly simplistic to define MIT as more free.
MIT license is absolutely not 'more free' than the GPL.
In fact, MIT means you give up effective ownership and control. You lose control and contributions.
And what do you get for that loss of control? Exposure. Or, in this and many other cases similar, you get diddly shit. Some company paracitizes your code, sometimes even demands SOC questionnaires and 'do this bug NOW', and other abuse.
> Not everybody has to be chasing money in all their activities.
Talk about missing the point! This was all about money. It was about a job at the company where the code is being used in a production manner. And they didn't even bother to give an interview.
And not many of us are independently wealthy, and can do things that we want with no monetary care. And, most FLOSS devs aren't that. Instead, they're being used as unpaid stepping stones so some overvalued AI hypesquad can vibecode (or slotmachine programming) faster.
> The author said he was proud of this outcome and nervous at how widely his hobby project will be deployed. That sounds like the ambition of many open source authors and a win. Might never have happened with GPL.
That's where I hope the author relicenses as LGPL and proprietary, and doesn't give Anthropic any more free professional work.
And if it never would have happened with the GPL, gasp, they would have had to pay developers to create it.
And until I'm independently wealthy, I too will license AGPL. If you're making money on my stuff, I want a cut. Simple as that.
All of this is built on exploiting the open source movement. Delineating between closed source ventures and Free community efforts is just good sense at this point. If they're going to take they must give back.
In all seriousness, good work. Sorry about the rejection, but it reminds me of the story about the Homebrew guy getting rejected by Google[1].
Also, that discussion gets pretty mean. Didn't feel like I wanted to send people there. I just wanted to give the guy a pat on the back, and bring some humor into it. Been there. Sucks.
I do the work because I see it as payback for all the great open source software I use all the time.
Something that isn't brought up enough in the "rewrite everything in Rust" discussions is that the API guidelines explicitly recommend MIT/Apache to "maximize compatibility" (i.e., corporate friendliness, or developer and user exploitation): https://rust-lang.github.io/api-guidelines/necessities.html#...
Your project has been around for a while, but it's crazy to me that anyone still open sources anything under MIT (or similar) in the era of LLMs. Are they that confident in their job security? Are they already independently wealthy? Frankly, even a proper copyleft license is likely to just be ignored, or the code laundered through an LLM-assisted rewrite, by these companies. I prefer to just keep anything I can't sell all to myself rather than release it, at this point.
Which is why my position is GPL > MIT..
This has changed everything. AGPL and GFDL from now on.
GPL makes them share or pay to relicense, since you own the copyright. with MIT, they don’t need to ask. MIT just benefits big corps. GPL better protects the open-source spirit, and paradoxically, the ownership of your work.
One other model that can also work well is to dual license as GPL + commercial, so people who want to publish their work can use the GPL license but you can potentially fund the project from license sales to closed source users using the commercial licensing option. I see this a fair bit in the audio community I work within.
A number of other people contacted me with offers so it looks like there will be a happy end to the story :-)
If I was you, I would probably feel similar "you used my project, you probably want to hire me!"
But there's a logical fallacy there.
Your creation being useful to a person or company ≠ you being a fit to work with/for them full time.
Still, you deserved human eyes on the question from their side.
Andrew@gambit.us
~~Have you considered a copyleft licence like LGPL?~~ Answered in a sibling comment
I wouldn’t say that’s exactly the case. Not to denigrate the author or anything, but this library is a relatively minor part of what Anthropic is doing. It’s a UI manipulation library, specifically one that simulates keyboard and mouse inputs. While something like that is certainly necessary for the project in question, it’s not anything that couldn’t be rewritten in-house without too much difficulty, especially since they’re only using a subset of the platforms supported by the library.
I’m sure that working on this project has provided the author with expertise in this area that Anthropic could benefit from, and so in that sense it’s still a shame that they wouldn’t give him an interview, but that’s really all that can be said about it.
This is my experience, at every group I’ve been in. Extending the date a bit is much easier than involving legal for approving a new library.
The group I’m in now sunk a substantial amount of money into a lawsuit for a library that accidentally made its way in, so are now “No LGPL.” with some crazy loops and approvals required if there’s really no alternative (very rare). From their perspective, it’s cheaper and safer to rewrite than not be in compliance, unintentionally or not.
Google is, but not from AI.
Expecting a reward from open source software is a recipe for disappointment. I have contributed code to projects by companies that say I'm a mentally-ill household object. I'm not going to change the license of my open source projects to get back at them, because the collateral damage against entities that aren't evil simply isn't worth it. (It's also somewhat unlikely that the people working on NTP servers at Facebook wrote those policies, so...)
I've mostly stopped applying to the big companies long time ago (10+ years) precisely because I'd never hear back regardless of the match or the credentials.
The only exception has been JaneStreet — they've contacted me almost immediately after a cold application with a small cover letter about my interests.
Yet going the referral route, it's relatively easy to get an interview almost anywhere, even Google or Apple.
I would recommend that here. There's no reason why Anthropic couldn't use your talents! See if you can find a friend-of-a-friend who is there, and then do a phonecall with them.
Huh. I guess if you decide to make OCaml your company's primary programming language, you have to take what you can when it comes to devs.
I doubt anyone who works there is "take what we can get" calibre. They want to attract people who casually solve college-level math puzzles for funsies. So I imagine it's the opposite and if you get hired there, you're surrounded by people who are extremely accomplished.
A thing to consider, though, is ethical: they seem to have been involved in market manipulation. [0]
Oh, they ignored him. I am not sure if that puts the company in a better light.
IMO I think foundational projects that every single bigtech uses like ffmpeg should get on this licence yesterday. They would start getting millions because it still would be way cheaper than making it themselves in their bloated cost structures.
See the comment of Manly read in this section. Once the threat of payment approaches, you can just switch to a free fork. A single person can't really win a trial against a big, well-funded company.
Might as well hire the actual expert.
(I’m not an accountant!)
Would be hilarious to bury a clause like “Modified MIT license — head of HR must publicly announce any employment application rejections of the maintainers while wearing a chicken suit).”
The model probably has the lib in it tbh.
But that might be just my frustration from experiences.
To continue the devil's advocate: why bother with all of this, if the company doesn't have to and the OSS version is enough anyway?
I don't know about the author's approach to this matter, but if I would find out that a company is making a killing using my software and then that company would refuse to even give me an interview I'd probably stop loving doing what I do. Sure, the software is under MIT license and it was the author's choice to do so, but what's the point of doing it under such a license when you can't even count of it mattering in a resume? What's the point of providing free labor to a company with revenue in billions? If you look at the author's blogpost, the only benefit the author mentions is making the number of downloads go up and that's just pathetic.
I am reminded of an another, similar case with a library called "FluentAssertions". This library used to be free to use by anyone until the author changed the license and started charging money for commercial use. The author did that because he spend several year maintaining the library on his own time and dime and megacorpos like Microsoft wouldn't even bother to donate despite using it extensively. What happened afterwards was that the author got shat on by everyone on the internet for daring to ask for money. In the company I work for his library has been replaced with an another free fork at a incredibly fast pace. All that free labor and the author got dropped as soon as they fell out of line.
The worst thing is that it wouldn't probably take much to make the author of the library happy. Even if they weren't interested in hiring him they could still acknowledge him, talk to him a bit to maintain good relations, throw him a nice donation as a thank you and now it would be a nice, good PR story instead of an another reminder that corporations are just looking to squeeze out value out of all of us.
> I found out that Anthropic had an open position in the team implementing the secret, unreleased feature of Claude Desktop using enigo
Obviously, you've provided value to a company in a really in-demand area. It doesn't feel right to treat the contributors like this. Sadly, it seems that the companies have the power and the intent to just abuse and exploit
I don't have a solution. I am just expressing my frustration from the perceived injustice.
A competitor could hire the OP instead, get them to work on improving the software for a few years. Giving the competitor a major head start.
Worst-case scenario, the tool they are building doesn't work out and Anthropic has a pretty good developer to put on other projects.
An FOSS project is rarely production ready that is really free as in beer considering TCO. Especially for a tech company.
If this wasn't available for free, they would gladly pay for a programmer to create it. But if it's already free, they can use it as a starting point. Maybe they'd need to internalize/extend it. But the option of paying for the work already done is gone.
Do this for each npm dependency and you're looking at huge savings.
If every company using a library chose the former, then every hour of development would be paid for (from the perspective of the maintainer) and the cost would be spread out across all its users.
Sucks, but that's the reality of hiring (and getting hired) in tech in general.
So getting an internal reference and being highly qualified for something they need done isn’t enough. You need to also make it past the 20 years old gate keepers and their amateur hour hiring process.
OP still has a chance now, maybe not anthropic, even other competitors can come knocking.
If you’re in the inside, it doesn’t suck at all, it’s so much safer.
Hiring a new person, based on a few hours of interviews, and a resume half full of exaggerations and lies, is such a ridiculous gamble. Worst part is, if you realize they’re not a good fit, it’s sometimes incredibly hard to get rid of someone, more often not an option at all.
Imo it's totally plausible that something will be expensive & time consuming to create, even with LLMs, but still easy to fork outside current licensing restrictions with LLMs.
Is there any evidence of this happening? And any legal theory behind how it might have the intended effect? Training being fair use does not make AI a magical copyright-removal box.
Training on most of the data in the world is important. Rewriting a document you don't own is not important. Or depending on what level the objection is made at, stealing some random library is not important.
I can almost guarantee that they didn't even read that application / cover letter and auto-magically rejected it.
"the team doesn't have the capacity to review additional applications"
Zero effort. They probably didn't even realize the relevance of that specific application for that role. Unbelievable, I swear!
Idk it sounds plausible that OP might just have been late to the party, and applied when the recruitment process was at the final stages.
Im curious at which point ppl will understand its counter productive.
Obviously, for writing and sending said job applications.
The world is also full of totally delusional people who dreamed up the idea of using winzip to compress VRAM on the GPU, and now Anthropic will definitely hire them for $1M year for this genius solution, so better write up a glammed up resume and auto-send it once a day for any open position.
I might just be old but i really haven't felt like contributing to open source at all lately because i've bills to pay and kids to care for and taking time out of this just for the sake of enriching some billion dollar corp that will eat me and spit me out doesn't feel like a good investment for my time.
Sometimes i feel sad that it came to this but this is the place we're living in right now.
[1]: https://www.cdflaborlaw.com/blog/federal-court-grants-prelim...
It's bit more AI now and bit less boilerplate rejections.
If you want to get hired don't focus on skills to build useful things. Focus on psychology and charisma.
You don't understand. You need to meet the hard skills bar, but there are far more bars than the hard skills one.
I think a hefty share of people here fail to understand the fact that there is way more to hiring a candidate than leetcode.
I remember back in 2014-2019, it was hard and competitive to contribute to open source projects as they were tightly guarded. There are many projects that I use now in package.json that are looking for a maintainer. A complete 180 flip.
My guess is that real free open source will disappear in a few years and what will remain are open source projects monetized by some business somehow.
It’s a sad reality but that’s what the current people at the top have decided today.
Always always always try to get into direct contact with the actual hiring manager. Blog author had a friend of a friend let them know a relevant role was open. The correct move is NOT to blindly apply. It’s to ask for an intro to the engineering manager responsible for the role.
"Coffee" with the friend of a friend would he better strategy than a cover letter in that case...more work, but better strategy.
Because logically, getting hired requires demonstrating you are "the kind of person we want to work with." Being qualified on paper is not necessarily required.
“Coffee” is a way to meet “a company.”
A letter is not.
Even a letter of introduction.
A lot of people with education in management/business do go into HR, at least in countries I know, and it does not help. People with extensive management experience would help but they will only take more senior roles.
The other qualifications open opportunities interesting and well paid careers. How would you attract those people into HR?
I am not even sure it would help if you could.
I think the suggestion in the old management book by the guy who turned around Avis that you should have an old style personnel department to do admin and advice, and managers should have more involvement might be a way forward, but I am not sure it would work given the current level of regulation (in the UK anyway - I imagine most wester countries are the same). A lot of the function of HR is to avoid legal risk (e.g. fire people according to the rules, so go through the motions of warnings etc).
What it really comes to is that a lot of people love to micromanage everything. If you hire someone that has integrity and educational background in subject, he/she will warn you if the decision you are making will have consequences in the long run. If you have someone that does not have relevant education, that simply does not happen. The managers micromanage, those people receive salaries and if they step out of the line even when they are right, they are reminded that they do not have relevant knowledge in said department (law/economy). This in turn leads to a lot of people gaining something called shallow experience which then in turns leads those people to hire someone that des not pose the risk to their position further down the line.
The problem being in this case is that there are a lot of misses that happen when the HR is organized like that; from illegal hirings, not knowing key economic factors, not having a clue about the business itself, no clue about laws and procedures and so on. Which in turn does not really protect the company because the company loses both the money and employees.
I can understand though, perhaps in a work environment where management is unlikely to be able to retain high skilled talent, you may want 'low-profile' workers that aren't going to have as many competitors chasing after them...
If you get in somebody who is a star, however minor, that changes the equation, changes the dynamic. Now that person can have more confidence, can have more sway in the decision making. If the company wants to let them go, then they might post a message to their followers, riling them up, creating bad PR for the company. It's no longer a simple equation.
So it all comes down to the insecurities of the company.
The probably most simple explaination would be that for some roles you like to have someone that can be easier "shaped" into a certain role. Someone who is already successful may bring their own system of doing things. This is great if it is a good fit, but can produce frictions if it isn't.
The next thing is that if you apply to a mediocre position with overly amazing credentials, it can raise suspicions. Something must be wrong with you, maybe you got amazing credentials, but you are complicated to work with. Maybe you're looking for the mediocre job just because you think it will be a walk in the park, etc. There are legit reasons for this (e.g. "my partner moved to $TOWN for her career and I am looking for something to do here, and you seem like the best fit. I know I am technically overqualified, but I wanted to go back to coding for years now and this offers me a geeat chance to give it a go").
Of all the senior canidates we have rejected the most common issue was that they didn't offer a convincing explanation to why they chose that specific position. The worst one was talking about how it would be a relaxing position for them.
>The worst one was talking about how it would be a relaxing position for them
What's wrong with that? Can't you compensate being lazy with being efficient?
When parent poster says things like “low profile” it should be interpreted as cheap and doesn’t know their worth. Assume all hiring managers want the least qualified and cheapest possible employee that can still get the job done.
Not always true, but true enough to be useful and more true than hiring managers admit to themselves. I’ve been a senior involved with hiring for years because while I full don’t want to manage, I also never trust my manager to hire well. They have multiple mutually exclusive narratives they tell themselves about how they hire/manage. Not all of them are true, and sometimes not any are.
It dependents on the size of the organization a lot. However in general it's likely that the new hire is the most competent of them all, which would be an immediate risk for some of the managers (e.g being displaced)
There are other risks like burn out as you may read a lot of OSS contributors have — so when someone is hit by burn out it will be across the board not that they somehow will perform at their peak at job while burned out by coding on side.
The new person could show how unproductive they are.
This is such a US-centric cliche that it even reads as a parody. No, the man isn't keeping you down.
That's exactly right.
> This is cope and propaganda to discourage people from developing their own brand.
Not really "cope and propaganda" when it's true, is it?
It seems like they didn't even look at his application.
I ended up building my own head hunting firm specifically to address the whole pipeline. That helped somewhat but head hunting is its own very odd space. Full of inefficiencies and bias.
With any AI company, there are always limits you hit. Energy, compute, optimizations, inference, team resources, money, and all the flows to make it a company. HR is usually the one that gets the fewest resources.
Honest question: what leads you to believe they should change their mind?
https://www.insidetechlaw.com/blog/2025/06/workday-ai-lawsui...
https://www.forbes.com/sites/janicegassam/2025/06/23/what-th...
The "it rejected me" in the headline should have been "it didn't notice me".
It’s also curious the author is looking inside the app for proof their software is being used. If it’s MIT, mustn’t the license be included and available somewhere easier to verify?
Maybe we ought to go back to paying for proprietary software. A lot of people used to make money that way, ie by selling their own desktop app.
Automated systems, AI screening, and incompetent HR people are the bane of modern recruiting practices.
I guess at least they're dogfooding it?
The practical reality of distributing is mildly complicated, but there's now lots of good cross-distro options, and not having to deal with code signing everything makes some parts much easier than Mac & Windows. Ignoring that many users is fair enough for a startup or first MVP, but quite surprising for a company at Anthropic's level.
"Overall I am overjoyed enigo is used in Claude Desktop and I tell everyone who listens to me about it :P. It's so cool to think that I metaphorically created the arms and legs for Claude AI, but I can't help but wonder if the rejection letter was written by a human or Claude AI. Did the very AI I helped equip with new capabilities just reject my application? On the bright side, I should now be safe from Roko's Basilisk. "
I also felt like this way that did they just AI in their interviewing process?
And I have a special love towards open source.
And I personally might be happy too that a company is using my work ,but in the name of the holy licenses, Companies are just exploiting the free nature of this and the fact that it seems like not even a human looked at the person for such job, who created a library that they are using it for free...
I was thinking of creating some code in MIT license, but I am going to create a code of AGPL except if you sponsor me on github or a special one time license which can grant you MIT.
People might say that I am not fostering the open source community, but I am not giving corporations free labour so that they can be billionaires.
I once saw someone write a software with the exact same idea (AGPL + gh sponsor me to get MIT) and the people in HN were pitchforking him, that's the harsh reality of the world. People want absolutely free labour.
I think open source needs to ask, Have we become the modern peasants in the name of our altruism?
I once told some non-techie folk about some code I wrote. It did something super simple and wasn't that big. They were all asking why I didn't sell it and thought it was crazy I would give it away for free with the BSD license. It was 900 lines of code... For us, that's nothing but for an average person they just think "I built it, I'll sell it"
I am still in high school, so I was doing some question sheet that our teachers provided and there was an answer key but it had answers of everything. Now I don't know how other people approached it but I am really impatient and so I just open up answer key side by side but it reveals every answer.
So I firstly created an AI to ocr to card generator but it was an hit or miss and so I discussed it with my friend and he said that he used to use paint and somehow in his convuluted manner basically have a slider which would reveal answer...
I found it incredible and so I just created a single index.html that can do it. (Although vibe coded), Now I can't even think of monetizing such ideas when I realize that there are creators of some really incredible stuff and long convulated stuff and even they aren't sponsored so I have always felt that the scripts that I write or projects around such ~.5-3k loc. I just don't think of monetization.
I just don't know.. I like hacking stuff, I just feel more comfortable rebuilding stuff even if its mediocre if I feel like I can change it to suit my purpose better
I think that the only other industry that is gives as much completely free stuff might be research/science related, but maybe its due to the fact that computer are computer science too and thus related to academics.
I really just love tinkering with software and just the aspect of freedom that it can provide , but sadly, I find it just hard to really make money without being a job and such stories on which we are discussing, just makes me feel like I am kinda right.
On one hand we have 100 million payouts to researchers and on the other we have this, such disparity is kinda sad I suppose.
Out on the super, super far end of the distribution you may have things like paying for what is essentially 900-ish lines of extremely, extremely carefully vetted code for things like encryption, but that is very, very exceptional.
I've got a few open source projects on my GitHub that are in the 900 line range, and I know they're used in a few "interesting" places but I'm not crying about it because the simple truth is the commercial value of that code is simply $0. If I tried to sell it to the people using it, they would perfectly rationally just say no. I am abundantly compensated for it by all the other open source software I get to use.
And we often get the luxury to ask questions and receive answers (Issues) directly from the manufacturer (author, contributor).
And we need not much investment to set up our own factory... thus "materials" can be free and then we give away our product.
This sort of silliness is what you get when you run crucial business processes using AI instead of humans.
1) Is the hiring AI so incompetent that it did not realize it had a "S-tier pull" in the process and should have immediately prioritized the find?
2) Was the candidate's submission so bad that a reviewing AI couldn't even tell the massive relevance he had to their work?
I suppose, alternatively, Anthropic could just not really care about Claude Desktop enough to hire a specialist for one part of the stack. Perhaps they're looking for much more "full stack AI" who can do a lot. They have 350-400 total engineers, is that enough to hire a specialist for Claude Desktop?
I guess my question is: Did the AI fail, did the candidate fail, or did the AI work well and we just don't know the criteria it was succeeding in using.
This reminds me of better call saul where hamlin paid millions of dollars out of his own pocket just to not have Chuck working at HHM anymore. Which by his own words was the greatest legal mind he ever met. Sometimes people's principles work against you. And you don't have any moral ground to challenge that.
On the other hand, this attention might be working in your favor though.
This post can give you some visibility unless somebody sees it as frustration/negativity then they won't bother either.
aside of the core topic, best way to get a job these days is unfortunately either some elite job boards that work and both sides know why... or personal relations.
All the automatic HR/recruitment platforms is illness and i'm sure that's what victimized your genuine application there.
Here is the US BLS breakout of demographics by occupation category: https://www.bls.gov/cps/cpsaat11.htm
Having a free pass for doing evil stuff is what gave man their bad rep, should we now for equity give women a pass to become the new slave lords?
Why the need for the sexist addition of ladies? People of all sexes and genders in HR can be clueless.
Come on...
Does anyone here have experience with them, or knowledge about whether that description is more or less correct?
"we want to hire someone to work in this super secret feature that uses lib X"
"cool! I'm the author of lib X"
"no, thanks"
https://x.com/mxcl/status/608682016205344768?lang=en
In all seriousness though, the situation sucks. But there's still upside. Someone might reach out.
Okay, they were just busy doing work and didn't have any time to look at applications so they shuttered the JD and auto-rejected anyone in the pipeline. Seems reasonable
Andrew Tanenbaum of the MINIX fame was similarly surprised to find that Intel had quietly included the OS he wrote in Intel chips, making it perhaps the most widely used OS in the world. He seemed disappointed no one ever reached out to him to tell him about it [2]
[1]: https://lukesmith.xyz/articles/why-i-use-the-gpl-and-not-cuc...
It seems obvious that if Tanenbaum, or any open source project used a GP license in lieu of a permissive legally familiar license like MIT or BSD, the likelihood of the project being used in a commercial product would reduce to nearly zero. Intel would have used a different OS for their management engine.
I'm glad the GPL exists and believe the world is a better place because of it, but it feels like more and more it's salad days are in the past and the world has moved on.
The ops experience reminds me of the story of the maintainer of homebrew that despite widely being used at google was not able to be hired for a job there. It's disappointing and feels unjust, and I wish it was different.
> I found out that Anthropic had an open position in the team implementing the secret, unreleased feature of Claude Desktop using enigo.
where enigo is his input library. It's quite interesting that you chose to end your quote a few words before the end of the sentence.
I'm a developer with a project they use, so, I thought, for sure someone would review my resume after applying on their website. Nope.
After being ignored for a while, even having to get a Master's degree because no offers after a Bachelor's, I finally emailed a Director, who was previously a fellow committer at the project. People under him were not hiring at the time, but a recruiter from a different group has contacted me shortly, and I've had a 2-day flyout onsite arranged for two different positions, and had offers to join either one.
Most executives and investors just throw shit at the wall to see what sticks, imo. Then move on to the next place. That's why golden handshakes exist.
Use GPL or AGPL. It's the best thing we have.
Remember that companies like Microsoft spend billions on PR and their goal is to make you think what's good for them is good for you. This is rarely the case.
Just ask your friend for an intro.
To drive the development.
To prioritize some bug fixes.
The only projects with a permissive license, I am comfortable sending PRs nowadays are the kind of projects that will hardly enable a big monopolist to extract more rent from society while being covertly funded by the debasing of currency promoted by the FED via Cantillon Effect.
Now that said, I think there's an opportunity here for you. It's rare, but I've known people who spun open source like this into a revenue source, not directly from the company but from other ways. My advice (if you have the time/motivation/willingness): Embrace Anthropic's use here, even maybe reach out and help accomodate them for features! Offer to help get the Linux build working (you would be a personal hero of mine for that!). They may even hire you once they see that you're genuinely good and helpful. I have seen many times in the past (including personal experience) trying to hire an open source dev, and we discovered that they weren't actually very nice people, and in about half the cases they were outright arrogant, possibly even with a superiority complex. The beautiful thing about code is that it doesn't care about personalities, but rather "does it work." In a workplace though, personalities are (often) even more important than code. We almost hired one of the most brilliant open source guys I've ever seen - his code was like a work of art, and his domain knowledge was unparalleled. But he was an asshole to the HR person, was openly critical of our interview process (during the interview!), and had virtually no tact when talking to what would be his fellow coworkers. It's ok (and sometimes very good) to correct an interviewer if they ask or say something technically incorrect, but the way you do that matters. Don't immediately retort with stuff like "that's completely and utterly wrong" or "that's a really stupid way to solve that." Nobody wants to work with people like that. Of course I'm not suggesting that OP is that way, but rather trying to share some personal experience as to why a company may not be in a rush to hire an open source contributor. To be clear, that doesn't justify what they've done, but I do think remembering that Anthropic isn't just a giant monolith, but is rather made up of other humans that are just as human as we all are can be very helpful at understanding why things sometimes go the way they do.
On the license question, I'm a huge proponent of the GPL/AGPL (I would always suggest licensing end-user software this way), but I think for libraries like this MIT is usually the best option. Had you licensed it AGPL it may never have gotten traction in the first place, and you'd be worse off than you are now. It's possible it could have led to licensing, but having seen the inner-workings of big tech companies, I highly doubt it.
Apologies for the length of this. It struck a chord with me because I've had libraries used in a similar way (though at a much smaller scale) and had to do a lot of soul searching on it.
Regardless, awesome work! At a minimum you now have an awesome story to tell and you're part of a small club of people that have been successful enough to get hit :-D
The first is the issue of permissive licenses like the MIT license, that seems likely far beyond an appropriate license structure for today’s world and environment, I would even argue inappropriate since the .com bubble. Software and creating has changed a lot since the 1980s to such a degree that I don’t think even the originator and early supporters of permissive licenses would be supportive of…peoples work being used in critical ways to build two and three digit billion dollar corporations without any kind of reward or compensation. It’s an odd kind of peak dystopian hybrid of communism and capitalism, sacrifice of the self for the benefit of the very few.
I think it is at least time to discuss archiving things like the permissive MIT license (assuming it even makes any kind of difference at this stage) that are from not only a different developmental stage, different environment, but even a totally different country, society, nation, and world even.
The second theme of this blog post seems to be the absolute seizure of the… what should we call it?…resource allocation of people? I cannot recall right now, but I feel like this is the second blog post themed around someone core to some function of some big tech company being rejected by said tech company; and that’s in the backdrop of the cacophony of people dealing with all kinds of dystopian insanity in the employment/job market from fake/scam jobs, AI interviews, etc. The system seems to be totally breaking down to some degree, even if it is still limping along, as is evident by the massively downward revised job creation numbers over several quarters now. How do you “revise” jobs numbers from 139,000 to 19,000? Ignoring any political partisanship, “revising” an estimate downward by 86% is not just an “whoopsie”, it’s evidence that thins are broken, regardless of why or even how. They’re clearly broken.
I have approaching 0% confidence with anything related to Congress actually doing its job since it has effectively abdicated its cute role that provides it legitimacy, but discussing both of these topics in public can have a chance at forcing the muppets in Congress to address the issues, even if only for narcissistic and selfish reasons of being (re)elected to enrich themselves after they’ve gone back on their lies to get elected. And no, neither team is the better team; it’s all a con-job.
> I feel bad about my tweet, I don’t feel it was fair, and it fed the current era of outragism-driven-reading that is the modern Internet, and thus went viral, and for that I am truly sorry.
It's also possible he would have been hired if he applied for L-1. A lot of people get an ego check applying to Google where they're a senior staff engineer or a CTO at a small company and get an L5 offer.
> But ultimately, should Google have hired me? Yes, absolutely yes. I am often a dick, I am often difficult, I often don’t know computer science, but. BUT. I make really good things, maybe they aren't perfect, but people really like them. Surely, surely Google could have used that.
Care to provide links...
How can interviewers be such stupid, the fastapi creator had the MOST experience with it, he created it..
This guy got rejected by some automated system without even interview.
Yes, this. Sometimes I wonder if those coming up with the Homebrew example have any experience whatsoever with software development. I mean, sure the project is popular and surely doesn't hurt on a resume. But does it showcase any level of technical expertise or mastery? No, I'm afraid not. I would bet that the majority of software engineers would be able to put together an equivalent system in a week or so. Think about it, and pay attention to what are the system's usecases. It's hardly rocket science.
Without no knowledge of the details further than mxcl's tweet; probably any performance issues even on simple code, get infinitely multiplied when running at Google's scale, slogging the thing, on Google's dime. From what I've seen of him, mxcl is good at designing a really approachable product, and on running an open source project. But homebrew is really slow, even on the latests Macs, even for basic cases.
To me it seems then that he'd be more fit for a product owner/manager position than an engineering one, and that could be the root of his not-hiring.
Antrhropic: tl;dr kthxbye
I think the world already grew tired of rug pull tactics. If you want your reputation to go down the crapper with a lame attempt to shake down an end user, go right ahead.
I like that people blog about these experiences and enjoy the insights, but I think it's never good for the authors..