Komodor

Komodor was founded in 2020 by Ben Ofiri and Itiel Schwartz around a moment every engineer knows: something breaks in production in the middle of the night, and the on-call engineer wakes up to spend hours searching for the root cause across complex, distributed cloud environments.

Since then, the company has expanded its product into additional pain points in the infrastructure world, including Cost Optimization, and launched the Agentic Operations Platform for Production, a platform that enables SRE, Platform, and DevOps teams to build, run, and manage AI agents in a secure, transparent, and controlled way. The platform includes a self-learning mechanism that allows AI agents to improve with every run in terms of accuracy and cost efficiency.

Komodor’s customers include some of the world’s largest and most innovative companies, including Nebius, Cisco, Booking, Merck, BlackRock, Dell, NICE, Rockwell, Nscale, and others.

Every Employee Helps Shape the Company’s Next Chapter

Komodor is working on one of the most compelling questions in the AI world: how to make agents truly work in production environments at some of the world’s largest companies.

The company employs approximately 85 people, around 50 of whom are based at its headquarters and development center in Tel Aviv, and has raised approximately $90 million to date from investors including Pitango, NFX, Tiger, and Accel.

The organizational structure is based on small teams, with intentionally flexible boundaries between them. Every employee builds end to end, learns every day from highly skilled colleagues, and sees the impact of their work with real customers. New employees go through a structured onboarding process that brings them directly into the core of the company’s activity and allows them to take part in a period in which Komodor is redefining what the production environment of leading organizations will look like in the AI era.

Addressing a Real Need

Komodor’s platform addresses the needs of today’s enterprise production environments. AI coding tools are accelerating the pace of development, but all of these changes ultimately land in production, on the same teams that are already under significant pressure.

Most organizations are already experimenting with AI agents in order to keep pace, but few are successfully moving them from pilot programs into production. This preserves the existing problem in which applications generate more data than human engineers can absorb, even with the most advanced monitoring tools on the market.

At the same time, an entirely new AI environment has emerged, with hundreds of agents running across organizations without a single coordinating layer. Adopting the platform can save engineers months of in-house development and allow them to focus on developing and researching products that move the needle, rather than on maintenance and firefighting.

Technology Already Running in Production

Komodor’s AI layer was built on six years of experience solving production problems and on deep data about how complex systems at massive scale actually fail.

Today, the platform offers more than 50 out-of-the-box agents specializing in specific development and observability domains, proven workflows for incident response, cloud cost optimization and development processes, as well as capabilities that allow customers to build and operate their own agents alongside them.

Komodor addresses two major challenges: first, creating a network of reliable AI agents that are ready to work alongside human engineers in managing advanced production environments; and second, providing a stable infrastructure for running those agents and continuously improving them over time.