Staff ML Engineer, Agent Training & Environments
Core
Building RL environments, verifiers, fine-tuning pipelines, and evaluation systems to train and judge frontier AI agents.
Role type
Staff ML Engineer (Agent Training & Environments)
Builds
RL environments, verifiers, fine-tuning pipelines, eval systems, and training/serving infrastructure for frontier labs.
Domain
AI Agents, Reinforcement Learning, ML Infrastructure
Deliverable
production ML models
Required skills
Python, system and API design, production code with coding agents, tooling/CI/harnesses, SFT, RL methods (GRPO/PPO/DPO), reward design, compute-economics, debugging training runs
Preferred skills
agent harnesses, coding agents, multi-tenancy/sandboxing, distributed systems, ML infrastructure, data systems, frontier lab experience
Technologies
Python, Node.js, TypeScript, React.js, GraphQL, GCP, Kubernetes, MySQL, Spanner, PostgreSQL, Kafka, PubSub
Responsibilities
Design and run RL environments for agentic tasks; build verifiers and graders; develop fine-tuning pipelines; manage eval systems for agent trajectories; scale training and serving infrastructure
Seniority
Staff, hands-on IC with technical direction
