Mercor
Member of Technical Staff, Deeptune Environments
$160K–$500K
New York City · On-site · Staff · Full-time · Posted Aug 10, 2026
- Experience
- 2+ years of full-stack / backend engineering, including at least 1 year at a startup.
About the job
Build and operate scalable systems and tooling for AI labs to create reinforcement-learning training environments from real-world tasks and human demonstrations.
What you'll do
- Build simulated environments, agent-facing tools, task definitions, and verifiers.
- Design and operate backend infrastructure for environments at scale, including containers, orchestration, queues, and observability.
- Convert human data into reproducible training environments and translate research goals into working systems.
What they ask for
- 2+ years of full-stack or backend engineering, including at least 1 year at a startup.
- Fluency in at least one programming language; Python is primarily used.
- Working knowledge of post-training, evaluations, and reward modeling to collaborate with researchers.
- Ability to scope ambiguous work, coach engineers, drive execution, and remain hands-on in the code.
Nice to have
- Sandboxing or virtualization.
- Browser and computer-use automation, CI/build systems, or developer tooling.
Pay
$160K – $500K USD base salary
Tools
Python
AI focus
Build reinforcement-learning training environments and related systems for AI agents; collaborate with researchers on post-training, evaluations, and reward modeling. The role does not involve training models.
Model-written summary · the original posting is authoritative