NanoClaw Creators Launch NanoCo AI With $12m

The team behind NanoClaw, an open source, enterprise-focused AI agent harness built on OpenClaw, is forming a new company to take the technology deeper into the workplace. Their goal; give each approved employee a secure AI agent backed by a constantly refreshed library of that worker’s documents, communications and other job-specific context.

The new startup, NanoCo AI, is led by former Wix.com engineer Gavriel Cohen and his brother, Lazer Cohen, founder of tech PR firm Concrete Media. The pair told VentureBeat that NanoCo AI has closed an oversubscribed $12 million seed round led by Valley Capital Partners.

The investor lineup includes several high-profile strategic backers from across the modern software stack; Docker, Vercel, monday.com, Factorial Capital, and Hugging Face CEO and founder Clem Delangue.

NanoClaw has gained traction as an open source, MIT-licensed implementation of an autonomous AI agent harness tailored for enterprise use. Building on that, NanoCo AI plans to commercialize the stack for large-scale deployments while keeping NanoClaw itself open and enterprise-friendly.

The company’s strategy is to maintain NanoClaw as an MIT Licensed standard and layer commercial managed services and integrations on top, aimed at enterprises that need secure, production-ready deployments rather than DIY experimentation.

Rather than framing the product strictly as an automation engine, NanoCo AI is positioning it as a kind of enterprise “second brain” for workers: a secure, AI-driven “professional assistant” that is continuously updated with relevant workplace context.

Gavriel Cohen, now CEO of NanoCo AI, describes the one-to-one assistant model as the key value proposition. In an interview with VentureBeat, he referred to it as a “professional assistant” and framed the impact in terms of individual productivity arguing that if a single worker’s AI agent can make them two to three times more effective, the case for wider rollout across the organization becomes compelling.

A central idea in NanoCo AI’s approach is that each human worker gets their own approved AI agent, tuned to their role and fed with their work context over time. As employees share emails, documents or call notes with the agent, it builds up an ever-updating knowledge base designed to stay aligned with that individual’s day-to-day responsibilities.

Within that model, NanoClaw serves as the underlying open source harness for autonomous agents, while NanoCo AI focuses on secure enterprise deployment and management. The company is aiming at organizations that want the flexibility and transparency of open source foundations, but also need managed services that address integration, security and scale.

NanoCo AI’s backers reflect that focus on enterprise infrastructure, with participation from players that span containers, hosting, SaaS collaboration and the open source AI ecosystem. Their involvement underscores growing interest in agent-based systems that can plug into existing workplace tools while remaining controllable and auditable for IT and security teams.

With fresh funding and a clear commitment to keep NanoClaw open under an MIT license, the Cohens are betting that enterprises will want both: a standard, open agent harness and a commercial layer that turns it into a practical, secure “professional assistant” for every approved worker.