Mirendil
Member of Technical Staff, Post-Training, RL Environments
$300K–$400K
San Francisco · On-site · Staff · Full-time · Posted Jun 24, 2026
About the job
Build end-to-end data systems and execution environments that support reinforcement-learning training and improve production model behavior.
What you'll do
- Build and automate data-collection pipelines for complex, long-horizon RL tasks.
- Develop systems to identify and prevent reward hacking, and scalable sandboxed environments for realistic tasks.
- Assess how training environments influence production behavior and collaborate on environments that improve it.
Pay
$300,000–$400,000 USD base salary
AI focus
Build data pipelines and training environments for reinforcement learning, including work to improve production model behavior and prevent reward hacking.
Model-written summary · the original posting is authoritative