- James Hawkins, CEO of PostHog, shares insights on their first 3 hires
- The first hire, Marius, was a full stack engineer hired as a contractor after seeing his work on Hacker News
- The second hire, Eric, was also hired as a contractor after applying through a remote job board and impressing with his skills and autonomy
- The third hire, James Greenhill, was a data engineer who starred their repo on GitHub and was recruited through a coffee meeting
- PostHog's transparency and open source nature attracted talented and proactive individuals to join their team

- The company prioritized being open source and transparent, attracting a specific type of person.
- The third hire came from reaching out to GitHub repo stars and the fourth hire was a graphic designer found on Twitter.
- The company focused on good design to attract developers and differentiate themselves from competitors.
- The fifth hire was an engineer found on a Hacker News hiring thread.
- The company prioritized scrappiness and speed in their hires and valued individual contributors over managers.
- The company only hires when they are certain the person is a good fit and is quick to let go of employees who are not performing.
- The company pays a large notice period to employees they let go to maintain a high standard of performance.

In this episode of First Five, we talk with the Co-Founder and CEO of PostHog, James Hawkins, about the first five hires he and his co-founders made when they were first starting out. Along with host and Managing Director of YC Continuity, Anu Hariharan, James discusses how PostHog created their culture of transparency and the development of their company handbook which can be read in full here: https://posthog.com/handbook/getting-started/start-here. Read more about PostHog's first five employees here: https://posthog.com/blog/posthog-first-five.

Chapters (Powered by https://bit.ly/chapterme-yc) - 
00:00 - Introduction
00:43 - First hire
03:13 - Second hire
07:41 - Third hire
10:15 - Fourth hire
13:14 - Fifth hire
14:46 - Culture and skills

Apply to Y Combinator: https://www.ycombinator.com/apply/
Work at a startup: https://www.ycombinator.com/jobs

#startups #hiring #business #hr

Welcome to the first 5, where growth stage companies funded by Y Combinator look back on their first 5 hires.

I'm a Managing Director at Y Combinator and work with our growth stage companies.

I'm here with James Hawkins, the Co-Founder and CEO of PostHog, a suite of open source product analytic tools that companies can host themselves.

Post Hog was part of Y Combinator's Winter 20 cohort.

So James, can you tell us about your first hire?

Who was it and how did you decide to hire for that role?

Sure, so my co-founder and I were frantically trying to get ready for a launch on Hacker News.

We already had this kind of date in mind that was do or die, it's going on the internet on this particular day.

And we're trying to build this massive product on that day.

And we wanted to build kind of libraries that would make our, make Postdoc work with mobile, with web, with desktop, and with a bunch of other kind of, you know, a bunch of programming languages.

So we hired someone called Marius as a full stack engineer.

The way that, and we hired him as a contractor to start off with, so we're like, okay, we just want to find someone who can ship a bunch of these libraries really, really quickly.

We had a remarkably efficient hiring process.

So it just so happened that I saw someone on Hacker News who had launched a product a little bit like ours, but clearly a hobby project during the batch around the time that we wanted to hire and so I just wrote him an email got like hey this is you know hey we're working on something that's similar to your hobby project but we're doing it full-time Would you be interested in doing a little bit of paid work with us to build out some of these kind of SDKs and libraries that we need?

So it was just from watching him do some actual work with us, it was really clear that he was very talented And that's kind of why we thought, Hey, we should just get him in full time.

You know, there's no test, like getting them to do the real work, because that's a test, especially for the company.

But When you decided to email him and hire him as the first contractor, was that even an interview or did you just go by the hobby project and you just wanted him on board?

Yeah, it was literally straight into a project, which is kind of terrifying, but It was very de-risked.

We're only committed to doing a couple of days' work this person if it doesn't work out.

But their work spoke for itself really quickly.

So yeah, the hiring time was like 10 minutes to write an email and then we're basically

If you approached it as an interview, that is what some people do.

And in your case, you know, things were moving really fast.

And I often tell with startups, every first hire is different.

No 1 has the same journey to the first hire.

So this was someone called Eric and we thought, hey, that contracting thing we did last time works so well to make this feel obvious that we should just repeat it.

This was once we'd raised our seed round.

We had a bunch of stuff we needed to build.

We had tons of issues, loads of demand, so we're feeling good that we can afford to hire someone else.

We dropped, just put the company on a jobs board for people looking for all remote jobs.

This is, I guess, actually very important.

So from very early on, Postdoc has been all remote.

Since we finished the batch, We've been designed to work this way.

We're building a kind of very community-led kind of product where the community is a bunch of random engineers all over the internet.

And so we thought, hey, if we are kind of building as a community ourselves of engineers all over the internet, then it'll just kind of mean that we're working with our community rather than being in like a fancy office and then expecting kind of outsiders to come join us.

So yeah, we put ourselves on a great job board.

We put kind of quite a lot of effort into a company handbook.

So we wrote out how our whole company works publicly.

We were quite inspired by GitLab and that kind of way of working.

So we have a similar feel if you go through our handbook.

The actual kind of policies and how we work are very different.

But the thing we have in common is a ton of transparency.

And for Eric, but also everyone that shows up now, the fact we're willing to kind of write out how everything works gives them a really good sense of what it's actually like working here.

Kind of later, you know, we're really big on being asynchronous, giving autonomy, like actual autonomy to people.

We have a direct culture, So, but you can really feel that for real when all of your policies, all of your work, all of your code is happening in public in a way that's, could de-risk it, I guess, from like an external candidates perspective.

So yeah, Eric applied on a remote job board, came in, we did an in, we, my co-founder and I both interviewed him separately.

We both felt this person is extremely smart, we think they're going to be very scrappy, which is what we want, we're now today like fiercely kind of anti-manager-y and very pro kind of scrappy people.

And we thought, hey, let's like see if he wants to do like a paid week of work with us.

And then for that week, we would just like pick up real life issues that we want to work on.

He was supposed to start on a Monday, but over the weekend he'd already shipped a bunch of stuff.

So before he even started, we already knew that we were going to end up hiring this person too.

Again, though, this was like a really efficient process.

Like he applied for a job really high, he was probably the only, I literally think he may have been the first person we interviewed, but we got lucky with how skilled he was.

But the thing that we learned from him as a hire is how important it is to have scrappy and proactive people that can handle autonomy.

Like Eric was the kind of person that you can just mention to him at a really high level, like, I think we should do this.

And then the pull request is there the next day for whatever this is.

And he's figured it all out and shipped it.

Which makes him a joy to kind of work with.

And that was really obvious through the way he handled the contracting period with us.

And so this kind of, we created the concept of like a super week where we started doing this every time.

That made it really feel really easy to move forward with him.

You said you wrote down how you work at Post Hog.

You were transparent about the culture and the norms.

This document, did it exist before you hired Eric?

We did it from basically when we started the company.

Early on, frankly, it was probably because we wanted to look more mature from a user perspective.

We're like, well if we have like loads of pages of our website we'll look more mature.

And it also helped us think through our philosophy around what kind of company we wanted to be.

So yeah, it's like very, it's very extensive what it covers.

It covers how, when we decide to fire people, it covers how much we pay people.

And you had all of this before hiring Eric?

Not all of it, but we had an unusually large amount.

For example, I think we had transparent pay already planned out for people before he started.

So we had like more than like, we're probably in the top like 0.001% of companies from a transparency perspective at that stage.

Sure, so this was a guy called James Greenhill.

He was a data engineer at Uber before he joined Postdoc.

We found him because he starred our repo on GitHub.

So I used to go through every single person that starts our repo for the first, like a hundred stars.

And I would basically try and buy, especially if they weren't someone that I'd recognize.

I would try and buy them coffee in real life.

And James D was 1 of those, I was like, oh, someone from Uber has started our repo, maybe Uber.

Like I want to just learn from him, like have you used this in production?

If you haven't used it, like I wonder what, I want to learn like what stopped you from using it?

He was really passionate about the software and that single click on a star button, of which we now have like 7,700 of them, led to that kind of call.

And then he was like, Oh, I'm interested in working with you.

It would solve like we've self built systems like this before for reasons around data control.

So we put them through the same process we like I didn't you had this kind of cool with 1 of us the other 1, like, I can't remember if it was Tim or I 1 of us interviewed him first, then the other person interviewed him, and then we ran post-log super week with him and just repeated it.

And again, it was obvious, like we were learning tons from him around how to make the platform more scalable.

He immediately knew much more than us around a load of things to do with data infrastructure.

Yeah, it was like, kind of frankly, I would never have thought we'd be able to hire like a data engineer.

He's worked at like a ton of other really impressive places.

But we didn't kind of go outbound looking for him.

He was someone who'd used the project, or he thought it was cool on the internet at first.

So he was kind of already qualified as being interested, I guess, before we reached out.

But at this point, I literally think it was like, we'd interviewed 3 people and hired 3 people.

So again, it was a particularly efficient set of hiring.

It is still kind of terrifying how quick we were to make decisions, but putting people at this table, you could work with it again.

It was probably self-selecting, like being open source, being really transparent early on, like led to a certain kind of person being interested in us because we really stood out as a business.

I love what you said about, you know, you used to do calls with GitHub repo stars and you wanted to learn more from them.

And you know, you probably never thought that was going to lead to a hire and here you are, your number 3 hire came from there.

I think the next employee was a woman called Lottie.

She started as a graphic designer, she's starting to go into product design postdoc now.

We knew that we would want to grow quite bottom up.

We're very much a product that is built for a single person to adopt for free on day 1.

And so we're kind of like, okay, how are we going to grow really fast?

You know, what would make me as a developer, what would kind of block me from wanting to try out some random open source projects?

And there are lots of things that we're like, no 1 else, like we want to see.

There are some stars, like are other people using it?

But 1 of the components we felt was good design.

Like if something's really well designed, even if it's like, it gives you a lot of confidence that it's going to be worth sticking around for and like deploying it and integrating it and trying it out.

And so we, you know, at the same time, we also thought the thing that's different about postdoc compared to the rest of the industry we're in is we're the developer focused product.

So we really want to build this like stronger kind of, we want to brand around that user.

Everyone else has like blue websites that are super boring.

You're on their pricing page, there's no pricing information, it's talk to sales.

So we're going to like name ourselves after a hedgehog.

We're going to have like, just like a weird, unusual style, because we are the weird and unusual 1 that's joined.

Like we're not going to win by being the same.

So I thought, Hey, if I have a graphic designer working with me, I think we could build something really different.

That doesn't feel like a templated generic kind of thing.

We found Lottie, literally on Twitter, We happened to see a tweet from a designer.

Her portfolio was really weird and kind of cool.

We didn't like, again, it was really quick, but we just saw someone we liked.

We reached out to them through an unusual channel.

Again, a bit like doing like a GitHub star, like a Twitter direct message to like a graphic designers, probably like someone who's like, again, she was early in her career as well.

So, you know, it's like, we're trying to like get ahold of someone that runs design at figma or whatever at the time, but we could just see talent there.

She seemed really scrappy and I think we worked with her, I'm pretty sure we worked with her on the same basis, like we got her to do some actual work with us.

And to quote my co-founder, watching her do her job felt like watching someone perform magic to us.

Like we could just see, within like the first day she produced some artwork for our blog and we're like, man this is the kind of thing I wish I could do but never will be able to.

So again, that was like another really easy kind of hire.

I guess the core components where we reached out to someone in an unusual way.

We de-risked it by doing some actual work.

And that again made it easy to move forward.

So what's fascinating about this story is I actually see similarities to your first hire.

You actually saw her portfolio of work on Twitter.

So who was the fifth hire and how did you go about finding them?

He at the time was living in Poland, or still is in Poland.

He was an 18 year old who was still at school.

And he found us on, I think it was a Hacker News, like who's hiring thread, which is consistently a good place for us to find engineers.

And that's because our product is built for developers.

We have this open source approach, we're really transparent.

He had a successful side project that had over 10,000 users.

So he could, which kind of demonstrated, okay, he can ship products from front to back and generally do full stack.

He can think a little bit like a product person.

He's passionate, like, and passionate about development.

It was like 2 weeks before he finished school, not college.

Let's try and we'll do our super week and again, distribute the process.

But again, it would have been like he came to us because he's because we posted on Hacker News or something like we have, we would have interviewed a bunch of other people at this point.

Yeah, we just we were looking for the kind of characteristics that like Maris and Eric had and Lottie that made them successful and the process that had worked so well.

We're just kind of repeating that part, but the actual place people coming in from was quite varied.

So 2 things before we wrap up that I want to touch on.

1, it sounds like all 5 hires are still with post hoc.

But clearly, you had tested for the skills on the job.

And there was a little bit of self-selection through all the documentation you did.

1 thing you didn't touch on was culture, right?

There's a lot of emphasis on how do you test for culture.

I did hear a theme though, which you mentioned scrappiness.

So somewhere along the way, you're testing for scrappiness and speed, But can you touch a little more on how did you test for culture or did you even know what culture was or did you define it after the first 5 employees came?

Scrappiness and speed were like the most basic things for us to get right.

We're like, we're only going to win by being scrappy and fast.

We're not going to win by being polished and slow.

When we looked at these first few people working, the thing that struck us as particularly cool was everyone was quite anti being a manager.

Like everyone just wanted to do actual work.

So we're kind of like, okay, we should like really, really make sure we prioritize caring about individual contributors at postdoc.

So if you're a senior at postdoc, if you're managing other staff or you're not, it doesn't affect your pay.

We also will, if someone's basically interviewing, you know, 30 people, if someone's interviewing saying like, hey, I mainly want to manage other people, I would like to grow a team, they won't get hired.

We found that's kind of an anti-persona for what works for us.

Our fundamental belief there is even if you are a manager, who's the most inspiring person to work for?

And the kind of person that is cracking out brilliant work that you would love to be producing yourself, we felt was more inspiring.

Even if it feels kind of like an inefficient use of time.

You know, for me, for example, it's like, should I really be like writing blog posts, tweaking copy, you know, bucking around, but doing individual stuff.

I think the answer, like it feels inefficient.

Like I could just hire like 3 people at this time to do this.

But I think the reality is like, if I set the example of like I'll do individual stuff, I'll crack typos in our documentation, our merge pull request will be open for a long time if I can, it means other people follow that behavior and then all of the individual people who are doing all the actual work will be much more effective and they'll want, you know, and they're working with you rather than for you at that point.

So that's a subtle thing, but we're looking for an anti-manager personality quite often, but we're relying heavily on centralised management.

We've written everything down, we have a system for how all this stuff works.

The last thing I guess is we have a, and probably the most important thing we've got right with hindsight.

1 is only hiring when we're absolutely certain this person is going to be a good hire.

Every time you have this like slight negative feedback, like, oh, this person seems great apart from this 1 thing, maybe it'll be fine, or we'll just like coach them out of it, invariably ends up becoming like 10, like magnified 10X.

When once that person is there, we've got better at picking up the signals for what that could be for us, it was wants to manage a team.

The other thing we've done is we let people go very quickly.

If they're not working, we try to give ourselves the conditions to do that as easily and as fairly as possible.

So for example, we pay 4 months notice to people if we let them go, which is an unusually large amount.

Our reason for this is we want people to be producing the best work of their lives.

1 of the things I think that inspires a lot of our own team is working amongst a bunch of people that are also doing the same thing.

And so if we have someone who's kind of not performing at the same kind of level, they're not passionate, they're not engaged, or they're struggling to get stuff out and we feel like it's not going to be possible to get them to the level of the rest of the team, we will part ways with them at that point.

We will give them direct feedback of this, but we are really decisive.

I would say probably again much more so than other companies, but we're up front around that in our documentation, in the habit, We'll talk to people about it when we hire them if they ask about it.

But that kind of like, we will let people go.

We're not afraid of making painful decisions like that.

Has been really critical to keeping up the standard of the first few people who are amazing.

You've done such a thorough job of the first 5 hires and shared such amazing insights for the founder audience here.

want to work for PostHog, visit posthog.com slash careers.

If you're looking to be 1 of the first employees at a startup like post hoc, check out Y Combinators job board at work at a startup.com.

See all Y Combinator transcripts on Youtube

Most Transparent Startup Ever? This Is How They Hire