Sam Lambert started at the bottom of the stack, as GitHub's first database administrator. A decade later he runs PlanetScale, a company built on the same technology that once held YouTube together.
The job nobody sees is the database administrator. When a site loads, you don't thank the DBA. When it goes down at 3am, that is the person paged to bring it back. Sam Lambert took that job at GitHub in 2013, when the company was small and growing faster than its infrastructure could handle. He was the first one they hired to keep the databases alive.
He never really left. A decade later, Lambert runs PlanetScale as CEO and President - a database company whose whole reason to exist is the problem he spent his career staring at. His answer, boiled down: databases should just work, and they should never go down.
"I have been in databases for a long time," Lambert has said. "I was a database engineer by trade." That sentence explains a lot about him. He did not arrive in tech through the founder track or the MBA track. He arrived through the terminal - the person responsible for making sure the data survived.
At GitHub he stayed roughly eight years. He watched the platform scale toward tens of millions of users, and he watched its databases start to buckle under the weight. Along the way he moved from administering databases to leading the teams that built everything around them. By 2018 he was VP of Engineering, running infrastructure and platform organizations that grew to around 350 people. GitHub Actions shipped on his watch.
Ask him how the climb happened and he does not describe a strategy for getting promoted. He describes the opposite.
His advice on management follows the same grain. Do it for the right reason, or do not do it at all. "Become a manager if you care deeply about people and care about the results that great people can achieve," he told the Future publication. He warns against taking the job for influence or title, and thinks companies that let strong individual contributors stay individual contributors are doing something right.
The pivotal moment of Lambert's career was not a promotion. It was a piece of open-source software. As GitHub's databases hit their scaling walls, his team went looking for a way out and found Vitess - a system originally built inside YouTube to run MySQL at planetary scale, sharding data across many machines without forcing developers to rewrite their applications.
That last part is what hooked him. Vitess promised scale that was mostly invisible to the people building on top of it. His team moved GitHub's largest tables onto it, piece by piece - notifications, pull request statuses - and in doing so learned both how powerful the technology was and how hard it was to run well.
After GitHub, Lambert spent a short stretch at Facebook working on internet-scale video and traffic infrastructure. It was, by his account, colossal in scale but missing something. He wanted to be back in product, building things people touch. When he met PlanetScale's co-founders, Sugu and Jiten - who had come out of the Vitess team at YouTube - the pieces clicked. He invested, advised, then joined as Chief Product Officer, and roughly nine months later became CEO.
For someone who came up through infrastructure, Lambert talks a surprising amount about taste. He compares great software to the feeling of a well-made car door - the "soft clunk" when it closes. You cannot point to it on a spec sheet. It works, and beyond working, it feels right. He thinks the best products come from marrying technical excellence to that kind of refinement, and he is willing to obsess over details most infrastructure companies never bother with.
His leadership views are blunter. He does not chase being liked, and he has a line about it that he clearly believes.
There is a mindset layer underneath all of it. Lambert describes moving from pessimism to a chosen optimism - the idea that "you can choose your experience of the world," and that reframing an obligation as something you "get to" do rather than "have to" do changes how you show up. He even talks about manifestation, not as magic but as a way of nudging your own attitude and decisions toward what you want.
Before he was running a company, Lambert was quietly backing them. As a Sequoia Scout from 2017 to 2022, he wrote early checks into a cluster of companies that would go on to define modern developer infrastructure.
It is a portfolio that reads like a map of where developers actually build now. The same instinct that told him Vitess was worth betting a career on told him these tools mattered before the rest of the market caught up.
The phrase Lambert keeps returning to is "the last database you'll ever need." It is a big claim, and he knows it. Under him, PlanetScale finished its pivot from a hosted Vitess service to a developer-experience-focused MySQL platform, launched in May 2021, and has since shipped products like PlanetScale Metal. He has also started connecting the database story to AI, speaking on "Agents and Infrastructure" and what autonomous agents will demand from the systems beneath them.
He measures success in years and decades rather than quarters, and he is comfortable making unpopular calls to get there. For someone whose first job was keeping other people's data from disappearing, the ambition tracks. The DBA who used to get paged when things broke now runs a company built on a single promise: that they should not break in the first place.