Daytona Platforms, a three-year-old startup building cloud infrastructure designed specifically for AI agents, has raised a 60 million dollar Series B round at a 1 billion dollar valuation. FirstMark Growth led the round, which the company announced this week, and the new money brings Daytona's total funding to 91 million dollars — a rapid climb for a company selling one of the least glamorous and most necessary pieces of the AI stack: somewhere for agents to actually run.
FirstMark also led Daytona's 24 million dollar Series A in February, and Fortune reported that roughly 50 million dollars of the new round came from existing investors doubling down. The Series B also drew a striking roster of individual backers: Stripe's Patrick Collison, Shopify's Tobi Lütke, Perplexity's Aravind Srinivas, Zoom's Eric Yuan, Datadog's Olivier Pomel, Quora's Adam D'Angelo, who also sits on OpenAI's board, and Mercor's Brendan Foody. Strategic investors Qualcomm Ventures, Samsung Next and Okta Ventures joined as well, alongside firms including Alumni Ventures, Evergreen Fund and Torch Capital.
Daytona's product is the "sandbox": an isolated virtual computer that an AI agent can use to write and run code, browse the web and operate software, without touching a customer's own systems. The company likens it to giving every agent its own computer, with sandboxed versions of Linux, Windows and macOS, plus GPU capacity, all available on one platform through a single API. Because the environments are pre-loaded from snapshots, Daytona says a sandbox can be configured in about 60 milliseconds, while a conventional cloud machine can take from 30 seconds to several minutes to spin up. Sandboxes keep their state when paused, can be resumed where the agent left off, and can be copied so an agent can try several approaches in parallel.
The pitch rests on a simple claim about how agents differ from ordinary software. "Agents don't work like the long-lived services today's cloud was designed for. They spin up thousands of short, stateful jobs nobody planned in advance, and they need a cloud built for that," founder and chief executive Ivan Burazin wrote in the blog post announcing the round. He has also framed the product in headcount terms: companies will soon manage thousands of digital employees, and each one needs a safe place to fail, so that a single misbehaving agent cannot cascade into wider damage.
The customer list suggests the argument is landing where it matters. Anthropic, OpenAI, Nvidia, Cursor, Cognition, Meta and Cohere all use Daytona's sandboxes to train and test models, according to reporting on the round, and data company Snorkel alone has provisioned more than a million sandboxes on the platform. Burazin told Fortune that Daytona's revenue run rate grew 25-fold between January and September, averaging 50.9 percent growth a month — the kind of curve that explains how a Series B reached a unicorn valuation eight months after the Series A.
Daytona began by automating development environments for human programmers before pivoting to agent infrastructure, and the obvious question follows it: why can the big clouds not simply do this themselves? Burazin has said investors pressed exactly that point, naming AWS. The company's answer is that agent workloads — short, bursty, stateful and numerous — are different enough from the long-running services the major clouds were built around to support a specialist. Whether that gap stays open as the hyperscalers adapt is the central risk in the bet. For now, the 60 million dollars will fund the expansion from sandboxes into a broader agent cloud, and the market has clearly decided the category is real.