NanoCo, the startup behind NanoClaw, a fast-growing open-source artificial intelligent agent that’s secure by design, said today it has raised $12 million in seed funding.

The oversubscribed round was led by Valley Capital Partners and saw participation from a host of others, including Vercel, Monday.com, Slow Ventures, Clutch Capital, Factorial Capital and Clem Delangue, the Chief Executive Officer of Hugging Face Inc. It came as the company announced the launch of its first enterprise-grade professional agentic artificial intelligence assistant.

NanoClaw is a secure and lightweight professional AI agent that runs locally on people’s computers, offering a safer alternative to complex agentic frameworks such as OpenClaw. It’s designed to serve as an enterprise-grade agentic assistant that goes beyond what simple generative AI assistants can do.

The agent learns an employee’s role and adapts to their work style through ordinary conversation. It slowly develops an understanding of the work each user does, the tools they use, the projects they’re working on, and the people they collaborate with, so that they can automate many of the tasks they perform with an unparalleled degree of accuracy. The user can then assign work to their assistant as necessary in order to become much more productive.

Because NanoClaw agents are deployed directly onto the user’s computer or a company’s own infrastructure, they can use Docker containers to isolate agent sessions to ensure that they’re only allowed to access files and directories they’re given explicit permission to open. They also integrate with Vercel Inc.’s ChatSDK and OneCLI’s credential vault, creating a unified architecture to ensure they cannot go rogue.

The way this works is that every request is run through a secure gateway that injects the user’s credentials at runtime, enforcing company-defined policies, based on the secrets management and vault systems used by the organization. Any sensitive actions require explicit human approval, and compliance teams will have enterprise-wide visibility into what each NanoClaw agent is doing.

NanoCo co-founder and Chief Executive Gavriel Cohen (pictured left, alongside his brother and co-founder Lazer) told SiliconANGLE that countless early adopters, including multiple CEOs have reported how NanoClaw helps make them two-to-three times more productive once they get the hang of using it.

“They’re asking how to roll it out to their teams,” he said. “They’ve figured out where the value actually lives: an agent has to be able to work inside the most sensitive parts of a business. Their email. Their customer records. NanoClaw was the version they could safely use themselves without trading security for functionality. NanoCo is allowing them to give that same unlock to everyone across the organization.”

Since launching in February, NanoClaw has achieved almost 29,000 GitHub stars, and has been adopted by executives at some of the world’s biggest technology companies, including Amazon.com Inc., Google LLC, Accenture Plc, Meta Platforms Inc. and SentinelOne Inc. It has even gotten an endorsement from Singapore’s foreign minister, Vivian Balakrishnan, who uses it to automate his own sensitive work and recently described it as having a “second brain” that he doesn’t dare to switch off.

“You get the output of a team two or three times your size, without expanding headcount until it’s time,” said another fan, Matt McConley, senior digital product manager at Johnson Health Tech and the founder of Looksphere.ai Inc. “After seeing what this does for a team of six, it’s obvious that every employee at every company must have one.”

Valley Capital Partners’ Steve O’Hara said AI is shifting from a tool that answers questions to become an assistant that can get work done for every business worker. “NanoCo has taken the lead, with formal partnerships with Docker and Vercel and a security-first architecture built in from day one,” he said. “Enterprises aren’t waiting another year to deploy. They’re doing it now with NanoCo.”

Image: NanoCo