Teams at Bespoke Labs

Build and scale pipelines that generate, grade, verify, and QA reinforcement-learning environments and agentic coding tasks. Create realistic coding worlds around production codebases, automate task creation into the thousands, develop throughput-enhancing infrastructure, and manage the full task lifecycle. Analyze failures, prevent reward hacking and grader loopholes, run workloads on GCP, and use coding agents to accelerate environment development and validation.

The Research Engineer will develop and deploy reinforcement learning environments, collaborate with research labs, and ensure high-quality systems at scale.

Design and maintain the infrastructure for RL environments, focusing on execution, performance optimization, and production excellence while collaborating with research teams and clients.