IA en un minutoNewsBusiness and products
Daytona raises 60 million dollars and reaches a 1 billion dollar valuation with its cloud for AI agents
IA en un minuto newsroom · Editor: Jon Elgezabal
In 30 seconds
Daytona has secured 60 million dollars in a Series B headed by FirstMark and is now valued at 1 billion. Its business is lending AI agents isolated computers in the cloud, where they can program and work without putting other systems at risk. The company, whose customers include Anthropic, OpenAI and NVIDIA, has gathered 91 million in funding and will open jobs in Singapore, Croatia and San Francisco.
A company that gives AI agents a safe place to work is now worth 1 billion dollars.
Daytona, which builds cloud infrastructure for AI agents, has announced a 60 million dollar Series B round at that valuation. FirstMark is leading it through FirstMark Growth, and the individual backers include Tobi Lütke (Shopify), Patrick Collison (Stripe), Eric Yuan (Zoom), Aravind Srinivas (Perplexity) and Olivier Pomel (Datadog), among others. Samsung Next, Okta Ventures and Qualcomm Ventures come in as strategic investors. Altogether, the company has raised 91 million dollars. Its CEO and co-founder, Ivan Burazin, told Fortune that roughly 50 million dollars of this round is coming from investors who already held a stake.
Its platform provides safe, sealed-off computing environments in which agents run code, work with software tools and carry out tasks, and they can be started with a single API, as Tech.eu explains. They run on Linux, Windows and macOS, and in its announcement Daytona adds environments with GPUs. Burazin told Fortune that Daytona can get a new environment ready for an agent in 60 milliseconds; the same outlet notes that starting a virtual machine on AWS, Azure or Google Cloud takes anywhere from 30 seconds to several minutes.
Burazin set up Daytona in 2023 together with Goran Draganić and Vedran Jukić, initially to automate the development setups that big companies use to write software, and the company later rebuilt its product around agents. Some of its customers used to run this infrastructure themselves: Writer had its own platform until it went past 20,000 environments a day, and on Daytona it went on to create 300,000 in a single month.
Among its customers, it names Anthropic, OpenAI, NVIDIA, Cursor and Cognition, as well as Meta and Cohere. In its announcement, the company says its revenue has grown roughly 25-fold since January; according to Burazin, speaking to Fortune, that figure refers to its annualized revenue pace from January to September.
With the new funding, Daytona plans to take its platform beyond its first isolated-environment product, expand its computing infrastructure and hire in San Francisco, Croatia and Singapore, according to Tech.eu.
Why it matters · analysis and opinion
The round confirms that AI money no longer goes only to the models, but also to the plumbing around them. When an agent writes and runs code on its own, someone has to give it a separate machine where it can make mistakes without touching real systems, and the big labs would rather rent that piece than build it. The fact that model makers, the teams that train them and the companies building products on top of them all ask for it at once points to broad demand, not a single customer. It remains to be seen whether the established clouds will narrow the speed advantage Daytona boasts about. For anyone testing agents, the practical lesson is simple: none should work with permissions over production systems without an isolated environment in between.
Official source: Daytona · Written with the help of AI: how we make the news
