From Code to Community: How puf Inspires More People to Build More Apps

On this episode of Build to Succeed, David sits down with Frank van Puffelen— “puf” to most of the developer world—an industry veteran, ex-Googler, and one of the most familiar names in the Firebase and Flutter communities. If you’ve ever searched for a Firebase answer on Stack Overflow, there’s a decent chance you’ve stumbled onto one of his posts. With tens of thousands of answers and tens of millions of views, his work has quietly supported a generation of developers trying to ship real products.

‍

If you care about impact as an engineer—whether through code, mentorship, or community—this conversation will resonate with you.

‍

From Stack Overflow Legend to DevRel Icon

‍

puf started as a curious engineer, and discovered he loved explaining things just as much as building them. Eventually turned that combination into a career in developer relations. Along the way, he’s developed strong opinions about teaching, communication, and the role AI should play in how we build software.

‍

As he puts it himself:

“If there’s one thread through everything I’ve done, it’s this: I want more people to build more apps. Every new builder we empower expands what’s possible in tech.”

‍

From 8-Bit Curiosity to Firebase and Flutter

‍

Puf’s relationship with technology goes back to the 8-bit era: Commodore 64, Atari, ZX Spectrum. At first, computers were just fun machines to play games on, like they were for so many of us. But at some point, the curiosity shifted from playing the game to understanding how it was made.

‍

When he started writing his first programs, he had that classic early developer epiphany: You can tell a machine what to do, and it will do exactly what you said—not what you wished you’d said. That unforgiving honesty of computers hooked him. It pushed him toward studying computer science and then spending two decades working as a software engineer.

‍

The turn toward public-facing work came almost accidentally. In his spare time, puf wandered onto Stack Overflow and started answering questions simply because it was a good way to learn. He noticed the same problems kept popping up—asynchronous logic, NoSQL data modeling, unfamiliar data structures—and he liked the challenge of not only solving them, but explaining them in a way that actually made sense to someone else.

‍

Those explanations caught the eye of the Firebase founders. What began as a habit of answering questions online turned into a job offer and, eventually, into a role where he became one of the most visible faces of Firebase—on Stack Overflow, at Google I/O, and on stages all over the world.

‍

Teaching as a Superpower

‍

A big thread in the episode is the idea that great engineers don’t stop at solving the problem; they learn how to explain it. For puf, that distinction is where impact really multiplies.

“Great engineers don’t just solve problems; they explain them. When you can take a complex technical challenge and turn it into a story others can follow, your impact multiplies. You stop just building software—you start building understanding.”

‍

He treats explanations almost like you’d treat a product feature: something you iterate on over time. His now-famous way of explaining async execution—the idea that code doesn’t always run in the neat 1-2-3 order you see on screen, but often in a 1-3-2 pattern—didn’t arrive fully formed. It took years of trying different metaphors, watching where people got stuck, and refining the message until he could see that “lightbulb moment” in their eyes.

‍

Part of his process is to test explanations on people who aren’t deep in the tech, including his wife. If someone outside the niche can follow along, it’s usually a sign that the explanation is genuinely clear. And one of the quiet joys he talks about is seeing those explanations escape into the wild—when other people start reusing his analogies and wording, spreading understanding further than he ever could alone.

‍

For any engineer, it’s a powerful reminder: Once you know how to make complicated things feel understandable, your influence isn’t limited to the code you personally write.

‍

Finding a Voice on Stage

‍

What might surprise people is that puf didn’t start out as the confident conference speaker many know today. He describes himself early on as the typical quiet engineer, the one who’d rather sit in the back of a conference room than be on stage. Large events weren’t exciting; they were draining.

‍

That changed when he realized that his enjoyment wasn’t tied to being in front of a crowd—it was tied to being helpful. When someone walked up to him at a conference with a real problem, something clicked. He could answer their questions, unblock them, and send them away better equipped than before. That feeling turned speaking from a performance into an extension of the same thing he loved about Stack Overflow.

‍

Over time, he found himself increasingly on stage: live-coding, presenting new features, and explaining how things worked. He admits that public speaking still comes with a rush of energy he has to control, but the fear that many engineers feel around it just isn’t there for him anymore. And he’s quick to point out that this doesn’t mean he never makes mistakes. In fact, he’s embraced failure as part of the experience.

‍

He talks about live demos that didn’t work and how those moments became “shared debugging experiences” rather than disasters. He invites audiences into the process, asking them to shout out typos during live coding so they feel like collaborators instead of spectators. That simple shift turns the talk into something interactive, human, and far less scary.

‍

At the core of his speaking philosophy is one simple idea:

“Public speaking terrifies so many engineers, but here’s the truth: your audience chose to be there. They’re not strangers; they’re just friends you haven't helped yet."

‍

Once you believe that, the room feels very different. You’re not being judged; you’re being invited to help.

‍

Rethinking AI: A New Kind of Hammer

‍

Eventually, the conversation turns to AI and large language models (LLMs)—a territory that can quickly become abstract or polarizing. puf’s take is neither breathless nor cynical. To him, AI is powerful, exciting…and still just a tool.

“We’ve built this mythology around AI like it’s a sentient co-worker. It’s not. It’s a tool, a new kind of hammer. The question isn’t what AI can do—it’s what you’ll do with it.”

‍

He compares this moment to earlier shifts. Firebase made it easier for front-end developers to build full experiences without thinking much about servers. Tools like Visual Basic and today’s low-code platforms opened up app development to people who might never have called themselves “programmers.” AI, especially LLM-powered tools, feels like another turn of that same wheel.

‍

He’s very aware of the tradeoffs. When you make it easier for more people to build, you also make it easier for them to make mistakes. He saw this with Firebase when unprotected databases started showing up in the news. Still, he argues, the alternative—locking people out of building altogether—is much worse. The real challenge is education, not blocking access.

‍

What really matters, he says, is how you approach AI depending on where you stand. If you’re a beginner in a domain, AI might give you a partial, fuzzy picture of how things work—useful, but dangerous if you don’t know how to check the details. If you’re already an expert, AI can act as a kind of supercharged assistant: a faster way to type, search, and explore ideas you already understand.

‍

He’s critical of the way AI is sometimes marketed, especially when companies overpromise what it can do. Grand claims like “it can generate anything you describe” don’t just set unrealistic expectations; they also alienate the very people who could benefit most from using these tools thoughtfully.

‍

Underneath the critique, though, is optimism. He sees AI as yet another force that can help more people build more things—as long as we remember that a tool is only as meaningful as what we choose to do with it.

‍

A Simple Mission: More People, More Apps

‍

When you zoom out from the specifics—Stack Overflow answers, talks, videos, product launches, AI experiments—you can see a surprisingly simple throughline in puf’s career. He comes back to it more than once:

“If there’s one thread through everything I’ve done, it’s this: I want more people to build more apps. Every new builder we empower expands what’s possible in tech.”

‍

You see that mission in how he helps experienced engineers reason about async and NoSQL. You see it in how he champions tools that lower the barrier to entry for founders and non-traditional developers. You see it in his encouragement for engineers to share what they know, whether that’s in code reviews, documentation, talks, or community answers.

‍

He pushes back on the idea that some people are “not technical” and instead frames programming as a way of thinking—about data, about systems, about behavior—that can benefit a wide range of roles. Even if someone never becomes a full-time developer, learning to think more like one can change how they work.

‍

If you care about developer experience, DevRel, teaching, or using AI in a way that genuinely helps people build, this episode hits a lot of nerves in a good way. It’s not just about tools or career paths; it’s about what happens when you decide that your work isn’t just to write code, but to help others write theirs too.

‍

You can catch the full conversation with Frank “puf” van Puffelen on Build to Succeed and hear the stories, tangents, and insights that didn’t fit on this page—but might stay with you the next time you open your editor, answer a question, or step onto a stage.

‍

‍