Hiring your first 10 engineers
Great hires scale companies. Finding and closing them is an important skill every founder should learn.
Hiring at startups is easier said than done. Closing your first engineer can feel like an uphill battle. Your lack of profitability will make many pause. Then there’s the fact that that you’re competing with offers from companies with deeper pockets. But you can offer things that larger companies can’t: greater ownership, a sense of purpose and significant upside should you succeed in your mission.
Search for people who value these qualities. Look for people motivated by a mission, not by titles. Hire generalists who have something to prove, not specialists who have proven themselves. Keep an eye out for people unfazed by obstacles and chaos, they’re the ones you want by your side when the going gets tough. Your first hires will shape your company’s culture, so spend time finding the right people.
Sourcing is hard work
Start with referrals from friends, investors and people you’ve worked with. Warm introductions are by far the most valuable. If your direct network doesn’t pan out, look in communities that share interests. Coinbase found many early hires in cryptocurrency subreddits.
When all else fails, it’s time to roll up your sleeves and get to cold sourcing. People who’ve solved similar problems at larger companies are great candidates. When reaching out to someone you don’t know, it’s helpful if you can make a meaningful connection with them. Look for people who’ve given talks or written about something you’re working on, and ask them for advice. Find people who contribute to open source tools that your company uses. Use platforms like AngelList, which automatically filter for people looking for startups. When all else fails, resort to searching on LinkedIn.
Great engineers get dozens of emails from recruiters. Be genuine in your email, show them you’ve done your research, and make a case for why they can help build your company. Even if they’re not interested, build the relationship. They may recommend a friend, or change their minds a year from now. It’s important that someone at the company becomes a top-notch sourcer. Very few external firms have the quality and reach that you will need, and the good ones like to work with larger companies.
Invest in a good interview process
It’s critical that you’re measuring the right signals and minimizing bias at every step of the way. You’re also trying to put your best foot forward. Candidates will look to your interview process as a yardstick for company culture.
There’s no interview quite as thorough as asking someone to do the job you need them to do. But most companies don’t do this. Instead, they make you solve contrived problems in one-hour blocks to see if you can cut it. Larger companies like this, because it’s an easy process to scale up. Here’s where you can one-up them. They can’t afford to spend the time that you can on each hire. Use this to your advantage, and design a more thorough and engaging interview. I recommend a format like this:
- Screening call (1 hour)
- Take home problem (2–4 hours)
- Team Lunch (1 hour)
- Paid Work Trial (1–5 days)
- Closing Meeting (1 hour)
Your first contact should focus on getting them excited about the team and the mission. Ask insightful questions to understand what they’ve enjoyed doing, what they want to do, and what they’re good at. Use that to paint a picture of the impact they could have by joining your team before you go into interview mode. Ask them to describe something relevant that they’ve built, and probe with clarifying questions to see how deep their knowledge runs.
The second step is the take-home problem. Ask the candidate to solve a real-world problem you’ve faced at the company and send you their code. This eliminates the artificial stress of an in-person interview. Candidates can put their best foot forward, on their own time and in their own environment. Expect some great candidates to drop out because of the time commitment. This isn’t always a bad thing, and filters for people who are very excited about your company.
The work trial is the most unusual step but is incredibly effective at measuring the right signals. Ask the candidate to fix bugs and build features with your team for a day or two. Compensate them well for their time, and get a contracting agreement in place. There’s nothing like shipping code together to get a good measure of what someone is capable of. It’s also a great filter for interest and signals that both sides are willing to spend the time to get it right. If you have a great culture, this becomes a strong selling point. Be accommodating when scheduling, because people have commitments outside of work.
It’s also important to have a closing interview with a founder or engineering leader. Ask them if they have any concerns about joining the company, and address them head-on.
Reduce subjectivity in decision making
If you don’t provide clear structure for your interview, unconscious bias will begin to seep in. Avoid this at all costs. For each interview, define what success and failure look like as objectively as you can. Create a rubric which measures a candidate’s performance across 5–10 attributes. For each attribute, assign a score or a binary decision. Don’t build your rubric on some abstract notion of what a good engineer is. A useful rubric appraises skills that will add value to your company tomorrow.
Independent decision making is equally important. Interviewers should submit a scored rubric before any discussions about the candidate begin, no matter how casual. Once you’ve walked the candidate out the door, get everyone involved in hiring into a room. Have the team simultaneously vote on the candidate. Then go around and have everyone share their rubric, and explain their scores.
Great hires scale companies, bad hires sink them. It’s tempting to relax the bar when you’re under pressure to ship. But hiring the wrong person for the job in the early days is incredibly hard to recover from. It’s important to have someone — usually a founder — impose a high bar, and help the room arrive at a good decision.
The best way to do this is by asking clarifying questions, diving beneath the surface layer of feedback. If someone says the candidate solved the problem faster than anyone they’ve seen, dig deeper. Was it because they were effective debuggers or designers? Did they have domain experience relevant to the problem that others did not? Look for patterns that are consistent across interviewers and lean on rubrics to make the right call.
Be fair with offers…
Everyone should win if the company wins, and the right incentive structure is critical. If you’re not experienced with early stage hiring, don’t innovate here. Learn from people who’ve spent the time making mistakes and getting things right. Don’t worry about getting them perfect, but avoid structural inequalities. Get legal advice on all employment contracts.
Find out what a similar engineer at a top-tier company would make in cash and equity. Use this as your baseline for determining your offer. Offer them 60–90% of what they’d make in cash and a generous chunk of equity. A good rule of thumb is to offer 1% of outstanding shares for a senior hire, and 2–3x as much for very senior hires. If you’ve done your cap tables right, the percentage of outstanding shares scales well with the growth of the company.
Use equity to align individual and company incentives. I recommend the tried and tested 4-year vesting schedule with a 1-year cliff, and a generous exercise window. Coinbase was one of the first companies to offer a 7-year exercise window. This signals that we care about people who work here, even after they leave the company.
… and deliver them well
Be thoughtful about how you deliver your offer to maximize your chances. If the candidate is interviewing at other places, have an open conversation about it. Ask them if your company is their first choice (assuming that compensation was comparable). If the answer is no, you’ve got some selling to do first.
Once you’re the candidates top choice (or at least top 2), then consider timing. You don’t want to deliver your offer days or weeks before other offers to avoid recency bias. Be upfront about this, ask when they expect other offers, and tell them you’d like to deliver yours at the same time. Don’t hand out exploding offers, but ask the candidate to commit to making a decision in a reasonable time-frame.
Avoid strategies that are contingent on negotiation. You can only do this if you make fair offers and are transparent with your philosophy. That said, give them control over their offer structure. Let them trade up to 5% of their total comp from equity to cash or vice versa. Present them as three distinct options that they can choose from.
Hiring strategy should evolve with the company. If you interview at Coinbase today, you’ll be treated to a very different process. As the company has grown, our needs have diversified. We’ve tweaked our processes to match, with an eye to inclusion and efficiency. In a follow-up post, I’ll talk about some of the challenges we faced, and lessons we’ve learned from hyper-growth.
Thanks to Jeremy Henrickson, Brian Armstrong, Harj Taggar, Reuben Bramanathan and David Farmer for reading drafts of this.
