In my years as an engineering leader, I saw and experienced the challenges of management at nearly every level, from early at Facebook, through co-founding my own company, and last as CTO at Dropbox. I’m excited to share some of the lessons learned from those experiences in this first in a series of essays on how to succeed as a VPE/CTO.
The first and most important element of success, as obvious as it might sound, is building a killer engineering team. And to do that, you have to win at hiring.
Mind the Funnel
Many technical founders haven't really recruited before. It’s important to stress that recruiting is your job as a VPE/CTO. A great recruiter/sourcer can help, but ultimately in a small company it all rolls up to the top of the engineering management chain.
There are three main parts of the job: sourcing, evaluating, and closing. Think of recruiting as a funnel game. Every week, you have to work each part of the funnel.
Starting with sourcing—this is where most people struggle. There are many reasons for the struggle, but the root cause is because you cannot adopt an efficiency-oriented mindset around the task. You have to be willing to do a lot of speculative top-of-funnel prospecting in order to get warm leads/candidates. This mindset is exceedingly hard for engineers because the hit rate is low and it can feel frustrating. However, there really is no way around it.
I used to meet at least 10 engineers for coffee chats a week when I was VPE at Dropbox. About half of those might interview and maybe 1 out of 10 would result in a hire. Not great numbers, but that is how hard the funnel metrics were in 2012—and today, It's much harder.
It’s also important to emotionally reframe top-of-funnel sourcing not as 'what is best for my company' but 'what is best for the candidate'. Oftentimes, this would mean that I would introduce the candidate to other interesting companies. Good will pays off.
Lastly, are you okay feeling rejected? Most VPEs/CTOs have crushed everything in front of them. So sourcing can be emotionally tough because most of the time the answer is NO. Are you truly willing to put yourself out there? Because you have to be.
Moving on to evaluation. In my experience, most VPEs/CTOs and their teams are reasonable at evaluation, having either been an interviewer at a previous company or been interviewed themselves. It’s important not to ignore cultural fit, though I’ll cover that more below. The most important thing is to avoid a consensus mindset. A VPE/CTO needs to own the hiring decision after taking all available input into account. This might mean going against your interview team’s feedback if you have more signal. Ultimately, you have to own the decision. Learning which hiring anti-patterns to avoid can help.
Finally, closing, while hard, is reasonably well understood. You have to paint a compelling picture of where the product is going, the technical problems to be solved and also have the candidate experience the culture you are trying to create.
Rope in your other founders, your investors and your broader team. In particular, I think the CEO of the company should be playing a big role during this part of the process. Their job is to sell.
If you’re properly managing the recruiting funnel described above, you’re accomplishing the single most important part of the VPE/CTO role.
Hire for Culture Fit
A strong company culture, and especially a strong engineering culture, is an incredibly valuable asset at any stage company, but in particular early-stage. Facebook and Dropbox have reputations for very strong and vibrant cultures. This is not by accident. Both companies do a phenomenal job of hiring people who are strong culture fits. Here’s how my experience at both leads me to think about finding recruit-culture fit.
Too many culture/fit interviews boil down to: Do I want to hang out with this person? A more thoughtful (yet simple) methodology: In every interview, ask yourself whether you got positive/negative/non signal about a company value. How does that work in practice?
Facebook Company Value: Move Fast and Break Things
One high-signal way we evaluated fit with this value was whether a candidate finished their coding question in time. It didn't have to be perfect, it could have some errors, but finishing was key.
We also got incredibly valuable signal from which tradeoffs the candidate made in order to move faster. Did they make the right simplifying assumptions? Did they take the shortest path while still solving the essence of the problem? This is a great example of how refining fit evaluation also helps with capability evaluation.
Dropbox Company Value: Sweat the Details
The most important thing was to look for correctness and attention to the small details. Did the candidate go the extra mile or did they end up duct-taping something together? You’ll notice that the two values mentioned above are purposely orthogonal to each other. The point is that you can get pretty different signal based on the company values. You need to be thoughtful about what you are looking for.
Dropbox Company Value: We, Not I
We would ask the candidate to describe their proudest accomplishment. As they would talk through it, we would be looking to see if they were giving credit to their team, using words like 'We' vs. 'I'. etc. This was a remarkably good filter for whether someone thought of the world in a I-centric way vs a we/team-centric way.
Facebook Company Value: Focus on Impact
We would ask the candidate to describe their proudest accomplishment. We would be looking to see if they drilled down on why this accomplishment was significant. Did it have downstream (and hopefully measurable) impact on users/customers? Again, same question but with a very different focus on what you are looking for. One important thing: There are no right or wrong company values. Every company picks cultural dimensions that they believe are important for them to win.
Focus On Builders, Not Managers
If you are managing your hiring funnel and selecting for strong cultural fit, you will quickly find yourself facing a new question: how far can you go without a formal engineering management layer? The surprising answer is pretty damn far, which means your hiring should for a while focus overwhelmingly on the people shipping product.
A lot of software management centers around three things:
- Planning (think OKRs, annual/quarterly goals etc.)
- Communication, make sure the team is coordinated to achieve goals
- Team Management (Recruiting, Feedback etc.)
Modern tooling (Asana, Slack, Gem etc.) has made it vastly simpler to do each of the above in a horizontal, scalable way. A lot of traditionally necessary management bureaucracy is now being slowly augmented or replaced by software. This is a secularly good thing because it makes management a more fun job, makes teams more effective, and extends the runway available until your hiring focus deviates from builders.
This doesn’t mean that you don’t need managers. Like most other applications of AI/Software, humans are augmented, not replaced. The key thing you need managers/leaders for is to act as tastemakers, identifying what great looks like and figuring out what to work on.
Managers still need to identify great candidates, help to decide on the right technical strategy and determine who to place in leadership positions. The difference is doing it at much larger scale than before. The scope and effectiveness of managers has been vastly increased.
Which brings us back to the original question. How far can you go? In my experience, you can have roughly 30-35 engineers without formal management. You will likely want some Tech Leads before then but that is way simpler than a management layer. Especially early on, speed is critical, and your hiring priorities should emphasize getting product out quicker.
If you are delivering on hiring, you are doing the most important thing a VPE/CTO can do. It will remain key to your role for as long as you hold it. Mind the funnel, hire for culture, and focus on builders—that’s how you win.