Magic
Member of Technical Staff, RL Research & Environments
$275K–$550K
San Francisco · On-site · Staff · Full-time · Posted Nov 8, 2024
- Visa sponsorship
- Visa sponsorship is offered, if possible.
About the job
Build and operate post-training data, evaluation, reward, and environment systems to improve long-context model capabilities and user-facing behavior.
What you'll do
- Create and curate post-training datasets, including through synthetic generation, targeted collection, and self-play.
- Build evaluation frameworks, reward signals, and training environments; run experiments to measure capability improvements.
- Maintain reliable post-training pipelines and work with Product and Research to turn capability goals into measurable iterations.
What they ask for
- Strong software engineering fundamentals.
- Experience building or operating large-scale data or ML systems.
- Ability to design and interpret experiments measuring changes in model behavior.
- Track record of owning experimental or production systems end-to-end.
Pay
$275K - $550K USD base salary; $275K – $550K USD (type unspecified)
AI focus
Post-training and reinforcement learning work to improve model capabilities, including long-context reasoning, coherence, context use, and tool-augmented behavior.
Model-written summary · the original posting is authoritative