Summary:
CoreWeave acquires OpenPipe, a startup focused on reinforcement learning for AI agents, to expand its cloud services.
The deal aims to help enterprises develop customized AI agents using advanced self-learning tools combined with high-performance computing.
OpenPipe recently raised a $6.7 million seed round with backing from notable investors like Y Combinator and Google DeepMind's Logan Kilpatrick.
This follows CoreWeave's earlier acquisition of Weights & Biases, showing a strategic move to enhance its AI development platform.
Reinforcement learning is key for improving AI model performance, and the acquisition will provide more resources for enterprise-specific training.
CoreWeave Acquires OpenPipe to Enhance AI Agent Training
CoreWeave, a leading provider of cloud servers for AI model training, has announced its acquisition of OpenPipe, a Y Combinator-backed startup specializing in reinforcement learning for AI agents. This move aims to expand CoreWeave's offerings and capitalize on the growing enterprise demand for customized AI solutions.
Image Credits: Just_Super / Getty Images
Brian Venturo, co-founder of CoreWeave, stated, "Reinforcement learning is emerging as a pivotal force to strengthen model performance on agentic and reasoning tasks. By combining OpenPipe's advanced self-learning tools with CoreWeave's high-performance AI cloud, we're expanding our platform to give developers at AI labs and beyond an important advantage in building scalable intelligent systems."
The terms of the deal were not disclosed. OpenPipe, based in Seattle, recently raised a $6.7 million seed round in March 2024, with investors including Costanoa Ventures, Y Combinator, Google DeepMind's Logan Kilpatrick, GitHub co-founder Tom Preston-Werner, and GitHub Copilot co-creator Alex Graveley.
This acquisition follows CoreWeave's purchase of the AI developer platform Weights & Biases in March, highlighting its strategy to move up and down the technology stack. OpenPipe is known for its open-source toolkit called ART (agent reinforcement trainer), which helps enterprises develop tailored AI agents.
As more AI labs and startups focus on reinforcement learning—a method that rewards AI models for correct responses to improve performance on specific tasks—CoreWeave aims to provide the necessary computing resources. The integration will see OpenPipe's team joining CoreWeave, and its customers transitioning to CoreWeave's services, enhancing the company's appeal to both large AI labs like OpenAI and smaller enterprises.
Comments