How The Browser Company moves so fast: 3 habits you can borrow

The Browser Company of New York, the startup behind the Dia browser and its earlier iteration, Arc, moves at a pace that seems impossible from the outside. It’s easy to assume its team of 65 is fast because it’s small and AI-native, but that’s only a part of it. The bigger factor is their unique way of collaborating that they deliberately designed to keep work from getting stuck.

Most AI gains die in the gaps between people; the reviews, approvals, and handoffs that make up roughly 80% of knowledge work. Flood those gates with more AI output and the system chokes. Atlassian’s State of Teams research puts a name to this problem: the fragmentation tax. The Browser Company (BCNY) is what happens when a company removes that tax on purpose.

Hire for judgment, not just output

BCNY’s biggest speed lever begins with who they hire. They recruit “un-blockable” people, flexible generalists comfortable working outside their own specialty, and they evaluate everyone against a bar built for the AI era: Independence, Instinct, and Impact.

• Independence: can this person go two weeks without needing to be pulled out of a rabbit hole?
• Instinct: can they find the highest-leverage thing to work on without being told? (BCNY says this is the hardest quality to hire for, and the one they grow rather than screen for).
• Impact: do they ship the right things, not just ship?

AI then makes these versatile “lane-swervers” faster still. The cross-functional range that used to take years to build is now within reach of anyone willing to try, and AI tools give them the confidence to attempt work they’d once have handed to a specialist.

“We just can’t support anybody who’s not independent, mostly because we as managers are finding it’s so much more productive to tell AI what to do than to tell people what to do. If there’s ever a question, it turns out it’s way better for the manager to just be telling AI and doing it themselves. So we really demand that.”

— Hursh Agrawal, Co-founder & CTO

The takeaway: AI simplifies execution, but it can’t automate direction. In the AI era, judgment is your real speed multiplier.

Give small teams end-to-end ownership

Most organizations do not lose time because people are working too slowly. They lose it in the spaces between teams: review queues, approvals, context gaps, and ownership boundaries.

BCNY minimizes that fragmentation in a few notable ways:

• Context and feedback are shared openly. Constant dogfooding and direct, in-the-moment critique keep everyone working from the same information.
• Tiny pods own work end to end, often with just 1–3 people. Decisions are pushed down, not up — the smallest possible group owns each call, with one clearly named person on the outcome.
• Leadership sets one clear North Star each season — this summer’s was proving the business model, distilled into blunt mandates like “prove $5/day of value” — then hands pods full autonomy on how to hit it.

The result is that AI-generated output is less likely to pile up waiting for another team to absorb it.

“Really small working groups of people who are really individually empowered is the real secret sauce.” — Ben Cunningham, Design Engineer

The takeaway: Not everything about The Browser Company’s pod model may lift cleanly into a much larger organization. But you don’t need to copy the model wholesale to apply its lessons. If your AI gains are disappearing, the problem may be your handoffs rather than your tools.

Settle contested decisions by building, not debating

One of the clearest patterns surfaced in the research is that BCNY prefers prototypes over meetings. When an idea is contested, or there are competing directions to choose from, the team builds something testable rather than debating in the abstract.

That works because AI-aided experimentation is cheap, internal feedback loops are fast, and the product is heavily used by the people building it. In practice, this lets the team resolve decisions in days that might otherwise turn into weeks of alignment.

Consider how one of Dia’s marquee features came to be. A non-engineer spent a few hours prototyping an idea with AI and shared it internally. By Thursday, leadership had named it a bet. The following Monday, a small team kicked off. About a month later, Dia’s full agent experience shipped: new tech stack, new model, new security architecture. From one person’s prototype to a shipped, architecturally novel feature in weeks, because nothing sat in a queue along the way.

The same instinct makes reversing cheap. Features, processes, even the planning cadence are treated as prototypes — shipped to learn, killed without ceremony. For example, they moved from six-week cycles to quarter-long “seasons” when the shorter rhythm stopped fitting how they actually worked. They dialed back Dia’s report-generation feature after both users and the team pushed back. And when a bet doesn’t land, leaders own it in the open — one employee noted there’s “no finger-pointing I’ve ever seen at the company.” “Strong opinions loosely held” is how they stay aimed at the right target while AI keeps changing what’s possible.

The takeaway: For larger companies, this may be the most portable lesson. AI has flipped the economics of prototyping: often it’s now cheaper to build the real thing than to align on a description of it.

What leaders should take from this: redesign the system, not just the toolset

BCNY is one company, and much of what works at 65 people gets harder at enterprise scale.

But the core learnings are universally relevant. In the age of AI, the bottleneck is shifting from production to decision-making. That changes what separates the fast companies from the stuck ones. The teams that get the most from AI aren’t the ones with the best tools — they’re the ones who are redesigning their hiring, team structure, and decision-making so faster output doesn’t get trapped inside slower systems.

AI raises the ceiling on speed. Your organizational design decides how fast you can go.