5 things I learned from 5 years at Vercel

july 2025 – lee robinson

After five years at Vercel, I just finished my last week. What a ride!

When I joined in 2020, we were 30 people and had crossed $1M ARR. Today, Vercel is 650 people and over $200M ARR. During that time, I went from IC to VP and worked on DevRel, product, community, docs, and more.

Here are some of the lessons I learned and how I grew as a leader and manager, as well as a bunch of behind-the-scenes photos.

Next.js Conf 2024

Vercel team dinner

Japan community meetup

Stanford hackathon

Lydia/Delba keynote

1. Go hard at work, then go home

I responded to a Next.js tweet while sitting on the beach during my honeymoon.

That probably sounds insane. Actually yeah, it is.

Vercel was the first startup I worked at. Before Vercel, I would leave work and turn that part of my brain off. But ironically, because of my passion for building things, I would spend my nights and weekends writing code, reading forums, and learning about the latest tools.

When I got to Vercel, I was able to blend my passion together with my work. I loved it. The same apps I hacked on over the weekend would also help me grow in my career.

But passion without boundaries will lead you to burnout.

If you're going to go hard at work and pour everything into it, you have to also learn how to "go home". The only way you can actually turn your brain off and go home is if you've built a support system at work. The engine must keep running.

Why was I responding to tweets on the beach? I hadn't built the system for others to take ownership. I liked helping developers. I liked turning upset customers into happy ones. So I was reluctant to "give away my legos".

But you can do both. You can build a self-driving car and still take the Ferrari out for a drive sometimes. I started to spend way more time on recruiting. I hired people I'd love to work for one day. And slowly I was able to build a system that allowed me to actually disconnect from work when I needed to.

I often see advice about work/life balance that overrotates into mediocrity at work. That's not what I want. I want to go hard at work. I want to ship lots of code, respond extremely fast, create high quality content, and more. There's no substitute for putting in the hours.

But when I'm offline, I'm offline. I'm 100% focused on my wife and daughter. To be clear, I still get this wrong sometimes. It's a never ending pursuit of work/life harmony.

Takeaway: You can work really hard and set clear boundaries.

After after party

NYC dinner with team

Me live in 8k

Trying to get the script right

My favorite swag

2. Everything can be done faster

It's your job as a leader to push the pace.

My apprehension about pushing came from the fear of being disliked. I didn't want to be perceived as an asshole. But after working for leaders who pushed me, and seeing the results, I realized being pushed is much better than mediocrity and stasis.

You can push the pace without being an asshole. The ideal state is that your team both loves what they're working on and can ship fast. Teams want to feel agency and durable ownership for what they're building.

If you're okay with things moving slowly, they will. Aggressive deadlines expose hidden complexity. They force conversations about what really matters. As a leader, you can speed things up by having these conversations and helping the team narrow focus.

My goal was to build a culture that set aggressive deadlines, but also was okay with sometimes missing dates. If there's no date, it might take 3 months. But if you set a date of 2 weeks and it actually takes 1 month, great, you just shipped a first version 3x faster.

I've watched radical deadline compression many times at Vercel. It often starts with a simple question: "what would it take to ship next week instead?" Extremely talented programmers find ways to get 10x more done in the same time.

It's one of my favorite parts about roles like DX and Design Engineers. These are folks who can code, design, build, and ship. They have extreme agency and operate more like founders.

Takeaway: Your greatest advantage can be how quickly you ship, listen to feedback, and iterate on the product.

rauchg speaking the truth

Delba and I doing the preshow

The crowd at Ship 2023

Vercel billboard in SF

DX team in 2023

3. Scale or die trying

CEOs have 1000 concurrent threads open, context-switching daily. They don't want to ping each thread to keep it alive. They need leaders who can own the thread.

Taking ownership doesn't mean doing all the work yourself. It means taking responsibility for the outcome, or if it can't be done as planned, communicating to reset expectations. To achieve this, you need to get really good at recruiting and hiring.

You must always be recruiting. If you don't scale the team, your best people get overloaded. If you get behind on hiring, you try to shield your team by doing work yourself. It's not fun.

There are only two hiring answers: hell yes or no. It means saying no to many good but not great candidates. Closing great people sometimes requires unconventional measures.

When the company is growing fast, there's always more to do than capacity allows. If you can't support another team's request because your team is too busy, you can't be upset when they find another way to get it done. You have to design an org that can say yes to the right things and no to everything else.

When someone isn't meeting the bar, you have to act fast. You want a performance and merit-based culture. At times, I felt I delivered feedback to my team early and with empathy. Other times I felt I waited too long. Those are the ones I regret.

When you delay giving critical feedback, you rob someone of the chance to improve for the sake of protecting your own feelings.

Takeaway: Scale yourself by hiring great people and building a support system.

DX team in 2025

The OG marketing squad

Design GOATs

Some of the best peeps

Joel and Kap, legends

4. Don't swoop and poop

Let's go back to 2022 when I was promoted to VP.

We were launching a new product. I started to review the plans and became frustrated. We could not ship this. Keep in mind, I hadn't been deeply involved with this plans to build this product.

The engineering team had been working on this launch for months. Then, I came in at the last minute and started asking a bunch of questions. I poked at decisions made months ago.

I didn't have the context. I didn't understand the constraints. I just disagreed and expected people to listen to me because of my title (ouch).

You can probably guess what happened next. Everyone was mad at me! In fact, there's a fun term for this: swooping and pooping. You drop in without context, make a decision, and leave others to clean up the mess.

I made a number of mistakes here:

1. If this was something I cared about, I should have been involved from the start. My favorite leaders are very hands-on, contrary to the "hire good people and get out of their way" advice.
2. If I wanted to change the direction, I needed to first establish trust with the team. I had to give them the same data and anecdotes I had. They should want to change the direction because it was objectively the right decision, not because I said so.

There's a better way that is obvious in retrospect: talk to people first. Shocking, right? You must actually listen and incorporate their feedback. Changing plans then becomes uneventful. Everyone already knows, and they all felt like they helped make the decision together.

Great leaders build alignment before making announcements. They don't use their title to override others. There will still be times when people don't agree with the direction, like team changes, but they should understand why and have built up enough trust with you that they disagree and commit.

Takeaway: Listen to your team and let the best ideas win.

Some of the Next team

Behind the scenes

Lee/G @ Apple

Green room prep

Meetup badges

5. It's okay to change your mind

I reviewed every Vercel tweet for years. I thought I had good reasons for this:

1. Someone needed to be the person to "stop the line" from shipping. I'd test the product, check all the links, and try to read the post from a beginner's mindset. What would someone who has never heard of Vercel think if this was the first post they read?
2. You can't take back social posts. Sure, you could delete them, but it looked sloppy. Once the message gets out into the community, it's hard to unwind. People would tag me regardless and I'd have to go clean things up.

I became the bottleneck and there was friction getting posts out. I thought I was "doing things that don't scale" by making myself always available. But when I went on an actual vacation where I turned my phone off, I realized it wasn't working.

The process I built to ensure quality was slow and had lots of approval steps. If I'm being honest with myself, I was scared about how we would scale quality and caring. It was hard for me to admit I was wrong, because I felt principled in the reasons I created the process in the first place.

My fallacy was treating all social posts the same. As Bezos says:

"What happens in companies is that you have a one-size-fits-all decision making process, where you end up using the heavyweight process on all decisions, including the lightweight ones. Two-way door decisions should be made by individuals."

Most social posts were not actually one-way doors. If you mess up the tweet about a small feature, you could just repost it and everything will be okay. The teams working on these changes should have the autonomy to ship.

I realized I was wrong, so… I changed my mind. I gave other people autonomy and documented what good social posts looked like. Something interesting happened.

The quality didn't suffer like I feared, and others learned how to rise to the quality bar we aspired to. More people became invested in becoming better at socials because they had ownership.

Takeaway: Be willing to change your mind when presented with new information.

React Summit presentation

Next.js GOATs

Jam sesh with Lindsey

Reactathon keynote

tomo and rauchg

Following my curiosity

The only constant in my career has been following my curiosity.

I'd been a product engineer for many years prior to Vercel, teaching developers on nights and weekends. I didn't plan on joining a small startup. So when I was offered to join Vercel, I followed my curiosity and took the job. I optimized for my potential to learn.

Following curiosity also means knowing when a chapter has finished. I'm grateful and thankful to the Vercel team for the past five years. It's been one hell of a ride.

Startups are messy and imperfect, but rewarding when building something you truly believe in. And I'm a massive believer in Vercel. It's a special team of people.

I'll talk more about what's next soon. For now, a heartfelt thank you to everyone at Vercel and in the community who has been a part of this journey for the past 5 years 🖤

Rest in peace JD

Pizza on take 100 of filming

Vercel SF meetup

#1 Remix fan

Our amazing community