Each information employee wants a pc. Brokers will want hundreds of thousands.
Daytona, the infrastructure firm constructing programmatic, composable computer systems for AI brokers, immediately introduced it has raised $24 million in Collection A funding.
The spherical was led by FirstMark Capital, with participation from Tempo Capital and present traders Upfront Ventures, E2VC and Darkmode. The spherical additionally included strategic investments from Datadog and Figma Ventures. At the side of the financing, Matt Turck, Common Accomplice at FirstMark, has joined the Daytona Board of Administrators.
Whereas immediately’s cloud infrastructure was constructed for stateless, immutable manufacturing workloads, it was not designed for the inventive, experimental nature of agentic work. Daytona gives a “new primitive”: the sandbox. These are programmatic, composable computer systems the place CPU, reminiscence, storage, GPU, networking, and the working system may be configured on demand, then began, paused, or snapshotted at any level.
“We imagine the following infrastructure shift is from human-centric cloud primitives to agent-native ones,” mentioned Matt Turck, Accomplice at FirstMark. “Daytona’s breakthrough is making ‘a pc for each agent’ sensible: instantaneous startup, persistent state, and the tooling brokers want to put in writing code, use Git, and execute safely at scale. That’s a foundational constructing block for the agentic economic system, and we’re thrilled to associate with Ivan, Vedran and the Daytona group.”
Daytona’s infrastructure permits an AI agent to launch a sandbox in milliseconds, discover a number of determination paths by forking into parallel branches, and snapshot mid-execution to make sure state persists throughout failures. This permits brokers to carry out complicated duties, from code execution to reinforcement studying, at a scale and pace people by no means required.
Additionally Learn: AiThority Interview With Claire Southey, Chief AI Officer at Rokt
The corporate reached a $1 million ahead income run price in beneath three months and doubled that determine six weeks later. Daytona’s clients vary from early-stage Y Combinator corporations to Fortune 100 enterprises, together with LangChain, Turing, Author, and SambaNova. Major use circumstances span code execution, pc use, and reinforcement studying.
“We’re thrilled to double down in Daytona,” mentioned Kevin Zhang, GP at Upfront Ventures and Daytona board member. “The group’s relentless tempo and obsession over developer expertise have been actually inspiring to witness, greatest mirrored by the fixed exercise and buyer love within the Daytona Slack and X. It’s only the start of this new agent period infrastructure alternative, and we can not consider a greater and extra centered group than Daytona to take it on.”
Daytona’s quick problem is maintaining with demand. The corporate says it’s hardware-constrained and can use the funding so as to add capability and broaden into new areas. Hiring can also be a precedence, Daytona is a 20-person group immediately. The remainder will go towards gross sales and advertising, the place the corporate has taken an old-school method: meetups, hackathons, and conferences in San Francisco, the place most of its clients are nonetheless clustered.
Additionally Learn: The Physics of Intelligence: Can AI Programs Develop an Inside Mannequin of Actuality?
[To share your insights with us, please write to psen@itechseries.com]
