Join Fulcrum

AI agents are going to write most of the world's software. We want to make sure humans can understand, control, and trust what they build.

Fulcrum is building the infrastructure and interfaces for human-AI collaboration in software development. Today's coding agents often wander away from user intent and produce code that is hard to maintain. Smarter models might locally improve outputs, but the fundamental bottleneck is human oversight capacity. We need infrastructure that lets a small amount of human intent scale to steer increasingly capable agents.

Who We Are

Our founding team has published research in top-tier ML venues like NeurIPS and COLM, gotten thousands of GitHub stars, won international olympiad medals, and founded MIT's AI alignment group. We've worked with frontier labs, data companies, and AI safety orgs, building tools for debugging AI agent evals and RL environments.

Hiring

We're currently hiring for a member of technical staff and a generalist to join our team. Reach out to team@fulcrumresearch.ai if you have any questions about the hiring process or about the company.