Senior Research Scientist
About the role
This team is tackling the challenges of long time horizons, context maintenance, and reliable operation across complex tasks by building a new generation of foundation models designed for persistent state, long-context reasoning, and agentic workflows. You'll join the research team focused on post-training, reinforcement learning, and model behavior, directly influencing how these models learn after pre-training to improve reasoning, planning, evaluation, and long-horizon performance at scale.
Responsibilities
- Research RL techniques for LLM pre/post-training and reasoning.
- Design large-scale experiments and evaluation frameworks.
- Improve model behavior across long-context and agent-style tasks.
- Train and evaluate foundation models ranging from 20B to 300B+ parameters.
Requirements
We're looking for researchers who have worked on challenging problems such as:
- RL for language models.
- Post-training and model evaluation.
- Large-scale experimentation.
Ideal candidates have helped push models beyond benchmark performance and care about their behavior in real-world settings.
Benefits
- Access to substantial compute resources.
- Collaboration with a highly technical research team.
- Freedom to explore ambitious ideas shaping future foundation models.
Pay
Up to circa $400,000 base salary plus meaningful stock (package negotiable depending on experience).
Schedule
Remote across the US or hybrid in San Francisco.