Bespoke Labs announced that it has raised $40 million across seed and Series A funding rounds. The company is building an AI agent training platform focused on environments that help agents become more reliable.
The Series A round was led by Wing VC, with participation from Mayfield, The House Fund, dbt Labs CEO Tristan Handy, and angels from Anthropic, OpenAI, and Meta. The company’s seed round was led by 8VC, with participation from Jeff Dean, Resolve AI CEO Spiros Xanthos, DevRev CEO Dheeraj Pandey, and others.
Bespoke Labs plans to use the new capital to expand its research team. The company also intends to scale its environment-building infrastructure and accelerate business momentum.
Bespoke Labs is focused on a key challenge in AI: agents can complete short tasks, but they often struggle to operate autonomously over longer periods. The company believes agents need realistic environments where they can practice, learn, and be measured against complex workflows.
The company builds environments that resemble real companies. These environments include large codebases, microservices, realistic logs, tickets, email, Slack, and other workplace systems so agents can learn long-horizon workflows tied to economically meaningful tasks.
Bespoke Labs also helps organizations measure and improve agents using techniques such as Genetic-Pareto Agent Optimizer, known as GEPA. This method automates prompt and policy search to help achieve higher accuracy faster than manual prompt engineering.
The company said customers use its infrastructure to train, evaluate, and optimize AI agents. Unlike companies that build simpler app-level environments using contractors, Bespoke Labs focuses on reinforcement learning environments developed through research techniques and collaboration with the academic and open-source community.
Bespoke Labs takes a research-first approach, with a team of research scientists and engineers. The company is a core contributor to Terminal-Bench, a widely cited benchmark for agentic capability.
The company is also behind OpenThoughts, an open reasoning dataset that has been downloaded more than 500,000 times. The dataset has been used by groups including Thinking Machines Lab, Meta, and Amazon.
Bespoke Labs’ research work supports its broader environment infrastructure strategy. This includes curation pipelines, methods for generating environments synthetically, and techniques for snapshotting real-world infrastructure and recreating it as a training environment.
Founded in 2024 by Mahesh Sathiamoorthy and Alex Dimakis, Bespoke Labs is based in Mountain View, California. The company is building infrastructure designed to help frontier labs and enterprises train, evaluate, and improve long-horizon agents for production.
KEY QUOTES:
“Frontier labs, enterprises, and all organizations relying on reliable agents need access to high-quality environments. This is the critical piece needed to optimize and develop agents. That’s why we are focused on doing research around environments, building the infrastructure around environments, and building the environments themselves.”
Mahesh Sathiamoorthy, Co-Founder and CEO of Bespoke Labs
“We work with labs on research programs spanning reinforcement learning, environment curation, benchmarking, and long-horizon modeling so that environments keep pace with the frontier of agent capabilities. I’m proud that we are a research lab pushing the frontier on environment infrastructure and curation.”
Alex Dimakis, Co-Founder and Chief Science Officer of Bespoke Labs
“As frontier labs and AI-native enterprises push the boundaries of long horizon agentic capabilities, a new generation of data and training infrastructure is required. Mahesh and Alex deeply understand the needs of leading AI researchers, and are building Bespoke Labs on a unique research-driven foundation to help advance and scale the state of the art.”
Peter Wagner, Founding Partner of Wing Venture Capital
