To apply, send us your resume and anything else you'd like to [email protected]

About Amigo

We're helping enterprises build autonomous agents that reliably deliver specialized, complex services—healthcare, legal, and education—with practical precision and human-like judgment. Our mission is to build safe, reliable AI agents that organizations can genuinely depend on. We believe superhuman level agents will become an integral part of our economy over the next decade, and we've developed our own agent architecture to solve the fundamental trust problem in AI. Learn more here.

Role

As a Researcher in Reinforcement Learning at Amigo, you'll develop our evolutionary chamber approach to continuous agent alignment. Working as part of our Research team, you'll create systems that enable agents to evolve under carefully designed pressures that align with organizational goals. Your work will focus on building efficient RL frameworks that optimize the integrated Memory-Knowledge-Reasoning (M-K-R) cycle, selectively targeting high-value capabilities rather than applying RL broadly. This research is essential for establishing the path from baseline capabilities to superhuman performance while maintaining strategic resource efficiency.

Responsibilities

• Design and implement reinforcement learning frameworks that enable targeted capability enhancement within the unified M-K-R cycle
• Develop techniques for RLVR (Reinforcement Learning with Verifiable Rewards) that use outcome-based rewards verifiable by external oracles or defined success criteria
• Create systems for iterative distillation and amplification that systematically improve capabilities through cycles of amplification and distillation
• Research methods for self-play reasoning that allow agents to generate their own tasks and learn to solve them
• Develop frameworks for continuous alignment that incorporate real-world interaction data and targeted simulations
• Design reward functions that align precisely with enterprise-specific metrics and success criteria
• Create multi-stage learning approaches that build upon our dynamic multi-step context engine and comprehensive evaluation system
• Research techniques for balancing exploration of novel strategies with exploitation of proven approaches
• Implement strategic resource allocation methods that enable different confidence requirements for different domains
• Collaborate with Simulation and Verification teams to create integrated evolutionary chambers