We’re hiring a full-time engineer to help build Prime Agent :)
Own features end to end across agent harnesses, cloud agents, and self-improving loops -- integrated with our sandboxes, evals, and hosted training stack.
Full-time, in person in SF:
Working on this with the team has been an absolute pleasure!
Prime Agent combines RLM-native tool calling, programmable context, and multi-agent orchestration to create a general-purpose coding harness that can improve itself. It also happens to score 95.5% on ARC-AGI-3.