Kilo Code Raises $8M Seed Round to Scale Open-Source AI Coding Agent
Kilo Code, a San Francisco–based startup building an open-source AI coding agent, has raised $8 million in seed funding as it accelerates development of its agentic engineering platform and expands adoption among software developers. Founded in 2025, the company aims to streamline AI-assisted software development by giving engineers faster, more flexible tools that integrate directly into their existing workflows without vendor lock-in.
The seed round was led by Cota Capital, a venture firm known for backing foundational software and infrastructure companies. The round also included participation from Breakers, General Catalyst, Quiet Capital, and Tokyo Black, reflecting broad investor confidence in Kilo Code’s approach to open, model-agnostic AI development tools. The funding provides Kilo Code with early capital to scale product development and support a rapidly growing user base.
Kilo Code was co-founded by Scott Breitenother, founder of Brooklyn Data, and Sid Sijbrandij, co-founder and executive chair of GitLab. The founding team brings deep experience in developer tooling, open-source communities, and enterprise software, which has shaped Kilo’s focus on reducing friction in AI-powered coding. The company positions its platform as a way to eliminate what it describes as “AI drag,” the inefficiencies developers face when switching between tools, managing model limitations, or dealing with opaque pricing structures.
Since its launch, Kilo Code has seen strong early traction within the developer community. The platform has been downloaded more than 750,000 times and reached the number-one position on OpenRouter, a marketplace for connecting AI agents to large language models. Kilo Code reports that its system processes more than six trillion tokens per month, demonstrating significant usage at an early stage. The platform provides access to more than 500 AI models, allowing developers to choose models based on performance, cost, or task requirements rather than being locked into a single provider.
Kilo Code is designed to operate across a wide range of development environments, including VS Code, JetBrains IDEs, command-line interfaces, and cloud-based workflows. Its feature set includes AI-assisted code generation, parallel agents capable of handling multiple tasks at once, managed indexing, cloud agents, and one-click deployment. By supporting both individual developers and engineering teams, the company aims to serve use cases ranging from rapid prototyping to production-grade application development.
The funding arrives amid intense competition in the AI developer tools market, where startups and established players are racing to define how engineers interact with large language models. Many existing tools impose constraints such as rate limits, downgraded models, or restrictive licensing terms. Kilo Code differentiates itself by remaining open-source and model-agnostic, giving users transparency and control over how AI is integrated into their workflows.
Investor participation in the seed round reflects confidence in both the market opportunity and the company’s execution. Cota Capital and the participating firms have pointed to the growing importance of agentic software development, where AI systems actively collaborate with humans to design, write, and deploy code. As organizations increasingly rely on AI to boost engineering productivity, tools that emphasize speed, flexibility, and openness are attracting heightened attention from venture capital.
With the $8 million in seed funding, Kilo Code plans to accelerate its product roadmap, expand engineering and community support, and deepen integrations with AI model providers and development platforms. The company’s focus remains on delivering what it calls “Kilo Speed” to engineering teams, enabling them to build and ship software faster while maintaining control over their technology stack. As AI becomes a core component of modern software development, Kilo Code’s early funding and adoption position it as an emerging player in the evolving landscape of agent-driven coding tools.
