Daytona has raised $60 million in Series B funding at a $1 billion valuation, as demand accelerates for computing environments designed specifically for autonomous AI agents.
The round was led by FirstMark Growth, bringing Daytona's total funding to $91 million.
The financing attracted a group of technology founders and executives, including Patrick Collison of Stripe, Tobi Lütke of Shopify, Aravind Srinivas of Perplexity, Eric Yuan of Zoom, Olivier Pomel of Datadog, Adam D'Angelo of Quora and Brendan Foody of Mercor.
Qualcomm Ventures, Samsung Next and Okta Ventures also participated, alongside investors including E2VC, Upfront Ventures and Pace Capital.
Founded in 2023 by Ivan Burazin and his team, Daytona originally focused on simplifying development environments before shifting toward infrastructure for AI agents.
Burazin previously founded Codeanywhere and the Shift developer conference and served as chief developer experience officer at Infobip.
Daytona is addressing a computing problem created by the rise of autonomous software.
Traditional cloud infrastructure was largely designed around applications, services and human developers. AI agents have different operating requirements.
An agent might need to write code, run tests, access a browser, interact with desktop software or execute commands inside a temporary computing environment.
Thousands of agents may begin working simultaneously, each requiring its own isolated resources.
These workloads can also be unpredictable. Agents may need to start within milliseconds, pause while preserving their state, resume later or create multiple parallel copies to explore different approaches.
Daytona is building what it calls an Agent Execution Cloud to support those requirements.
Its platform provides isolated computing environments, or sandboxes, that agents can use to execute tasks without obtaining unrestricted access to a customer's infrastructure.
The company supports Linux, Windows, macOS and GPU-enabled environments through a common platform.
That allows AI developers to build agents capable of interacting with different types of software, including native desktop applications, browsers, development tools and GPU-intensive workloads.
Daytona also provides capabilities for persistent state and branching environments, allowing agents to preserve work and create separate execution paths.
Its commercial traction has accelerated considerably.
According to the company, revenue has grown approximately 25-fold since January 2026, while average monthly net revenue retention has reached approximately 135%.
That metric suggests existing customers are significantly increasing their spending, although Daytona has not disclosed absolute recurring revenue or profitability.
Its customer base includes Meta, Cohere, Writer, Clay, n8n, Postman, Snorkel, Surge AI and several companies developing AI training environments.
The company reports that Writer previously managed its own infrastructure for more than 20,000 daily sandboxes before migrating workloads to Daytona.
Writer subsequently provisioned approximately 300,000 sandboxes through Daytona in a single month.
Another customer, AfterQuery, replaced an internally managed Google Kubernetes Engine environment with Daytona.
These examples illustrate the company's proposition: specialized infrastructure can be more practical than building and maintaining isolated execution environments internally.
Daytona expects the need for sandboxes to grow as AI agents become more capable and operate for longer periods.
The same infrastructure could support software development, computer-use agents, automated testing and reinforcement learning environments.
However, the category is becoming increasingly competitive.
Cloud providers and specialized startups are building agent execution infrastructure, while model developers may eventually integrate more of these capabilities directly into their platforms.
Daytona must demonstrate that its performance, security and operating costs remain attractive as workloads expand.
Its business also depends on customers choosing independent execution infrastructure rather than relying on the environments bundled with their model providers.
The reported growth in usage and customer spending suggests early product-market fit.
The longer-term test is whether Daytona can become a standard execution layer for autonomous software, capturing a durable share of the infrastructure spending generated by AI agents.