Cognition AI Eyes $40 Billion Valuation as Devin's Revenue Races Toward $1 Billion

Cognition AI, the company behind Devin — the world’s first autonomous AI software engineer — is reportedly in early-stage talks with investors for a new funding round that could value the startup at $40 billion or more. If the round closes, it would represent a more than 50% jump from the $26 billion valuation Cognition achieved just three months ago in May 2026, when it raised over $1 billion. The new raise, first reported by Bloomberg, is said to be targeting at least $1 billion in fresh capital.

The Numbers Behind the Hype

What makes the $40 billion figure plausible — even conservative, some would argue — is the trajectory of Cognition’s revenue. In May 2026, the company disclosed that Devin had reached an annualized revenue run rate of $492 million, up from $37 million just one year earlier. That is a 13x increase in twelve months. According to sources cited by TechCrunch, the new round is being benchmarked against Cognition hitting a $1 billion annualized revenue run rate, which would mark yet another doubling in a matter of months.

Enterprise usage of Devin has reportedly grown 50% month-over-month through the first half of 2026. The company’s customer roster includes blue-chip names like Mercedes-Benz, NASA, Goldman Sachs, and Santander — organizations that pay for Devin’s ability to autonomously plan, write, test, and ship production code inside existing development workflows.

Perhaps the most striking data point: Cognition has said that approximately 90% of its own code is now written by Devin. The AI software engineer is not just the company’s product — it is its most prolific employee.

From Competitive Programmer to AI Empire

Cognition was founded in November 2023 by Scott Wu, a former competitive programmer who won three gold medals at the International Olympiad in Informatics. Wu previously co-founded Lunchclub, an AI-powered networking platform, before turning his attention to what he saw as the defining challenge of the era: building an AI agent that could do real software engineering work end-to-end.

The company launched Devin in early 2024 to enormous fanfare and immediate controversy. Early demos showed the agent independently navigating codebases, writing functions, running tests, and deploying applications. Skeptics questioned whether the demonstrations were cherry-picked, but over the following eighteen months, Devin matured into a tool that large enterprises now trust with production pipelines.

Cognition’s trajectory accelerated dramatically with its acquisition of Windsurf (formerly Codeium) in July 2025 — a deal that brought in an AI-native IDE with $82 million in annual recurring revenue and a rapidly growing enterprise user base. That acquisition came on the heels of Google’s $2.4 billion acqui-hire of Windsurf’s original founders, a move that left the Windsurf product and team available for Cognition to absorb at what many viewed as a bargain price.

The Agentic Coding Gold Rush

Cognition’s potential $40 billion valuation must be understood within the broader context of an agentic coding market that has exploded in 2026. The competitive landscape now includes Cursor (the AI-native IDE that recently partnered with SpaceXAI on Grok Bot), GitHub Copilot’s agent mode, Anthropic’s Claude Code, OpenAI’s Codex CLI, and Meta’s Muse Code. Each is vying for a share of what analysts project will be a multi-hundred-billion-dollar market as software development increasingly shifts from human-written code to AI-orchestrated systems.

What sets Cognition apart is its focus on fully autonomous, end-to-end task execution rather than copilot-style assistance. While tools like Copilot and Cursor excel at augmenting human developers in real time, Devin is designed to take a high-level task description — “refactor the authentication module and add OAuth support” — and complete it independently, surfacing for review only when it encounters genuine ambiguity.

That positioning has resonated with enterprises looking to compress engineering timelines. The 50% month-over-month growth in enterprise usage suggests that what began as an experimental tool has crossed into mission-critical infrastructure for a growing number of organizations.

Valuation in Context

At $40 billion, Cognition would be valued at roughly 40 times its projected $1 billion annualized revenue run rate — a multiple that, while eye-watering by traditional software standards, is consistent with the premiums investors have been placing on AI infrastructure companies throughout 2026. For comparison, at the May 2026 round, Cognition traded at approximately 52 times its then-current $492 million run rate. The compression of the revenue multiple from 52x to 40x, even as the absolute valuation jumps by $14 billion, reflects both the company’s breakneck revenue growth and a mild cooling of the most speculative AI valuation fever.

The broader funding environment remains extraordinarily supportive of AI companies. In Q1 2026 alone, AI startups captured approximately $242 billion — roughly 80% of all global venture funding, according to Crunchbase data. Within that pool, agentic AI and coding infrastructure have emerged as two of the most heavily backed subsectors, driven by investor conviction that software development is among the first industries where AI agents will deliver measurable, enterprise-grade ROI.

What Comes Next

The reported talks are described as early-stage, meaning terms, lead investors, and final valuation have not been finalized. Bloomberg noted that the discussions could still fall apart or result in a different valuation. But the mere fact that Cognition is already engaging investors just three months after its last mega-round signals extraordinary confidence from both the company and its backers.

If Cognition does hit $1 billion in annualized revenue — a milestone it appears to be approaching rapidly — it would become one of the fastest companies in history to reach that mark, trailing only a handful of consumer technology platforms. For a B2B enterprise software company selling autonomous coding agents, that trajectory is without modern precedent.

The implications extend beyond Cognition’s cap table. A $40 billion valuation for a pure-play agentic coding company would further validate the thesis that AI agents are not a speculative category but an immediately deployable productivity layer — one that enterprises are willing to pay for at scale, today. It would also intensify pressure on competitors, from well-funded startups to Big Tech incumbents, to demonstrate comparable revenue traction.

For Scott Wu and the Cognition team, the bet has always been that the future of software engineering is autonomous. The market, it seems, is increasingly willing to underwrite that vision at a price tag that would have been unimaginable when the company was founded less than three years ago.