Replit CEO Predicts the 'Self-Driving Company' Will Reshape Work

By Daniel Okafor

Replit's support tickets no longer reach a human first. An AI agent triages every one. It scans GitHub issues, the codebase and Linear tickets before a person ever sees the request. When that person approves a fix, deploying a patch takes one click — and a second agent does the work.

Amjad Masad, Replit's CEO and co-founder, laid out this system at TechCrunch Disrupt in San Francisco on Thursday. He calls the operating model behind it the "self-driving company." He does not mean robotaxis. He means a company where agents do most of the grunt work while people supply judgment and taste.

"Our company entirely changed how we operate, and I would say the closest analogy is a 'self-driving company'," Masad said. "The company is sort of self-improving, so a lot of the work that's going on is driven by agents."

What changed inside Replit?

The shift did not happen overnight. Masad said his team spent the past year reflecting on how AI reshaped their own workplace. The turning point came earlier this year, when AI agents became more reliable.

The support-ticket loop shows the mechanism. An agent performs the initial triage across three systems — GitHub, the codebase and Linear. A human then receives the request with all the context already assembled. That human decides how to proceed. Deploying the fix launches another agent with a single click.

"You have an entire loop where agents did most of the work, but crucially, there's human judgment in the middle, and we don't think that's going away anytime soon. You need human judgment; you need taste," Masad explained.

The model mirrors how autopilot works in aviation. The machine flies the route. The pilot handles the decisions that matter.

Are agents taking jobs or changing them?

Masad's answer is the second option. Agents change what people work on, not whether they work.

"We're spending a lot more time figuring out what to work on," Masad said. "I go around the company, I see a lot more discussions, a lot more whiteboarding. We spend a lot more time on architecture, how to build things."

The day-to-day picture at Replit backs this. Less ticket sorting. More whiteboarding. More time on architecture decisions. Human effort moves up the stack — from executing tasks to choosing which tasks deserve execution.

What does Masad think happens next for workers?

He warned that the change has already arrived for some and not for others.

"The reality is, all of our jobs — perhaps everyone in this audience, especially — changed over the past year," Masad told the Disrupt audience. "But a big part of America and the world, they still don't know that that change is coming, and it's going to be surprising to a lot of people."

That gap — between companies already running on agents and workers who have not felt the shift yet — frames the risk. Masad did not specify a timeline for when the surprise lands.

Is there a positive case for AI at work?

Masad thinks so, and he argues Silicon Valley undersells it. When Replit talks with other businesses, he said, the conversations center on how AI helps people become more productive and happier at work. The happiness claim has a specific mechanism: agents make work less boring and more creative by absorbing the repetitive tasks.

"There's a positive vision of the world that Silicon Valley is not really putting out, and we're hoping to do that," Masad added.

Replit sits close to this shift. Businesses adopt its AI coding tools to build and launch their own software, so the company watches the workplace transformation from the vendor side.

For business leaders, the takeaway is concrete. Replit runs a loop where agents triage, humans judge and agents execute — and Masad frames that structure, not head-count cuts, as the real story of AI transformation. The companies that design that loop early will set the terms for the workers still waiting for the surprise.