HN user
marcell.ortutay at gmail dot com
Our founding engineering team is going to be capped at 3 or 4 max engineers, and to get 20% you need to take a pay cut. So far people have generally taken the higher cash offers. The token grants on those translate to 4-8%, which is still more than typical.
It also depends a lot on the person, I have a pretty high bar for "exceptional": https://steve-yegge.blogspot.com/2008/06/done-and-gets-thing...
We pay very competitive on the cash side (over 200k for exceptional engineers), and have offer options that are 10x more competitive on company stake side than other companies (up to 20% for exceptional engineers, but with lower cash as a tradeoff).
I challenge anyone to find better founding engineer compensation. Comparisons from YC: https://www.workatastartup.com/jobs/l/software-engineer
It's similar but slightly different:
1) It's not a security, so it's not a share of the company. In order to be compliant with US law, we can only offer a utlity token 2) Crypto is a 10x better underlying data structure and architecture for the financial system: https://ortutay.substack.com/p/the-computer-science-case-for...
Will fix it shortly, the CSS / Javascript on that is a little buggy. I wanted to make neat visual effect but it breaks in a lot of cases.
Well, they don't have to be employees. It's not for everyone, if you don't want a crypto token compensation, Fetchfox is not the right company for you to work for.
While I agree that 1% is low. 25% seems crazy?
(author/fetchfox ceo)
I usually make a few different offers that have cash / token allocation tradeoffs. Higher token = lower cash, and lower cash makes you more like a founder.
Just to give a concrete example, for a recent very good candidate, the offer on the high token side was ~20% of the expect total allocation and 150k cash. There is also a higher cash option with lower token allocation.
For less exceptional candidates, the offers are a lower token allocation the equivalent cash.
I've noticed that recruiters recently will use euphemisms for crypto/blockchain/etc
I've seen this also, and I think its lame.
This is why at Fetchfox I explicitly use the phrase "crypto token" in the article and in all offers offers to prospective hires. It's a crypto token, so we call it that.
There are a lot of negative associations with the term, but using euphemisms just confuses people and makes it seem like you're afraid or that you're trying to trick them. I say, call a spade a spade, and call Voldemort "Voldemort".
Why does no one run a completely employee owned tech company?
I thought about setting up Fetchfox this way, and in some ways we are an employee run company. Everyone including me gets the same crypto token as our stake in the company's success. I get a higher stake as the founder/ceo, but some offers give the employee 0.5 for every 1 of my allocations, which I think is pretty fair.
Long term, it would be nice if I had <50% stake in the project, and it self-managed somehow, similar to crypto projects like Ethereum.
That said, the phrase "employee owned" has some bad connotations. It has an implication that you are not trying very hard to grow, or that you are somehow less committed to company, or that you are some sort of co-op. For these reasons I don't like use that phrase.
(fetchfox CEO here)
I have a few pragmatic reasons why I want to use a crypto token, instead of traditional instruments.
The first one is liquidity timing. With a crypto token, you can have 24/7 trading and liquidity from basically day 1. There is no need for second markets with high fees and complicated process. Just a 10 second Uniswap contract.
Second, you don't need to "manage a cap table". The token trades and moves freely on the blockchain, permissionlessly for anyone who wants to use it.
I have a long article at https://ortutay.substack.com/p/the-computer-science-case-for... about why I think crypto rails are fundamentally better than the current banking system, from a computer science perspective.
The best example is how money transfers work in the current banking system. A money transfer is the canonical example of why you need transactions in a database: you want to deduct money from person A only of you successfuly transfer money to person B. But guess what: this is never executed as an actual atomic transaction, because A and B are usually at separate banks, in separate DB's.
Compare this to the blockchain: every single transfer is an atomic, universally auditable transaction. If you ignore all the noise and scams in crypto, this just makes 10x more sense to me as an engineer.
FetchFox as a company has two goals:
(1) Make the best possible AI scraper (2) Prove that crypto tokens are viable and better alternative to traditional company structures
It's lame that so much crypto stuff is caught up in scams that obscure the underlying value.
Apologies, just made it public: https://docs.google.com/document/d/1VvxEQBRexuFJT5qCr9MCeZRV...
Hey guys, CEO of FetchFox here. Happy to answer any questions about the crypto token. I'm also making the draft doc public, please be aware the details are still being worked out, but we will publish the full plans on our site within weeks.
Draft doc with comments enabled: https://docs.google.com/document/d/1VvxEQBRexuFJT5qCr9MCeZRV...
FetchFox.ai | https://fetchfox.ai | REMOTE (US) | competitive compensation
FetchFox is an AI web scraper. It convert takes a prompt and a URL, and crawls that site to give structured data.
[edited to remove listing, as we are working through existing applicants at this time]
It hasn’t been an issue yet, but I’m sure it will come up at some point. If you see a problem please file an issue.
I’m making a free open source library for this, check it at http://github.com/fetchfox/fetchfox
MIT license. It’s just one line of code to get started: ‘fox.run(“get data from example.com”)’
Would it be possible to do an OS level ad blocker that works similarly to uBO?
I used to work at 23andMe, AMA
What is the uBI status on Brave, Edge and Opera?
I checked the code and found the issue. It's a result of Gemini's larger context window.
Basically, the Foxtrot scraping library sends the page in chunks. The chunk size is capped at the max context length of each model, which for Gemini is 1,000,000 input tokens for lite. That's compared to 128,000 for GPT-4o-mini.
Typically, you won't need all the tokens in the page, and sending a million tokens when 100,000 will work is wasteful in terms of cost and runtime, and can also hurt accuracy.
I'm going to re-run the benchmarks with a cap on the prompt size for models like Gemini.
I’ll give it a shot and update the results.
Unfortunately the bigger issue with Gemini is cost, which is too high for the scraping use case.
Wow! If true then it’s different than when I was there.
The statement is pretty vague, and it’s hard to know what it means on a technical level.
I'm not really sure, probably far away. Most of the genomic analysis showed very weak correlations. I was more on software side of things.
Yes, the same code runs for all customers
The company bet heavily on pharma/genomics, and it was a bad bet.
When I was there, people were pretty confident in this bet. They had just signed a huge deal with GSK, so it seemed to be going well. There wasn't widespread dissent at the time (~2016-2017). I imagine its different now that the stock price has crashed over 10x.
The company did follow Ancestry.com pretty closely. Ancestry did not bet heavily on genomics. Instead, they bet heavily on a subscription model and focused more on consumer interest in their ancestors. This has worked out a lot better for them than 23andMe.
FWIW, I agree it's obvious in retrospect that pharma was a bad bet. Leadership should have made better decisions.
I didn't follow competitors too closely, but Color may be what you're looking for. I don't know if they sell direct to consumer though.
If you use 23andMe, and request data deletion, they will do a best effort to delete the data. It's part of their GDPR requirement. When I was there, I worked on this project and they put a lot of effort into it. A big chunk of engineering org focused GDPR compliance for a month or two. They definitely don't intentionally keep data around if you request deletion.
The one caveat is that data deletion is hard, and its possible that some gets accidentally retained. I left the company over 5 years ago, so I don't know how good their deletion process is now.
One final note: to keep costs low, 23andMe doesn't look at your whole genome. They only look at a handful of "SNPs" in your genome that are known to be significant. If you've heard how we share 99.5% of our DNA with chimpanzees, this is what they're basing this on. They look at the <0.5% of DNA that commonly varies between humans.
The reason I mention this is that, if you're very interested in your DNA sequencing, you may want to opt for a higher cost service that does full genome analysis. I don't know any names but I believe there are some DNA services that do this.
The company was not very well run so I’m not surprised. Their stock price has tanked over 10x since IPO, and it dropped by half in the employee lockout period after IPO.
Kind of, dev ops and big data
Not directly, afaik they never transferred the data.
However they sold access to the data to a bug pharma company (GSK). This was widely publicized. Not sure if that counts: GSK had some ability to look at the data but didn’t have an on premise copy of it.
Also, I worked on the GDPR deletion project. I can attest that they do best effort to delete your data when your request that. At least when I was there, this was the case. One caveat is for coding errors, oversights and bugs.
Software engineer
I did not, but that’s because I joined late and had a high strike price. I also turned down a second market offer to sell at around $15/share. Oops.
Many people sold during internal private liquidity events, and they did well. If you joined early you did well.
Yes but I was one of only a handful of people with this access. It was heavily audited. AFAIK no one ever used it inappropriately.