about 5 hours ago
San Francisco, CA, USAMid Level / Senior
Responsibilities
- Design and build reinforcement learning environments for real-world use cases.
- Create verifiable reward functions to evaluate model performance.
- Develop and maintain evaluation harnesses for task suites across models.
- Conduct post-training experiments to validate learning signals.
- Package environments and results for reproducibility and contribute to research publications.
- Help identify and define future use cases based on model weaknesses.
Requirements
- Master's or PhD in AI, Machine Learning, Computer Science, or a related field.
- Strong knowledge of reinforcement learning and LLM post-training.
- Proficient in Python with strong software engineering skills.
- Hands-on experience with LLMs and related evaluation methodologies.
- Familiarity with containers and infrastructure for reproducible setups.
- Ability to work independently and manage problems end to end.
- Willingness to relocate to the San Francisco Bay Area.
Benefits
- Ground-floor role with significant influence on team direction.
- Direct collaboration with founders and experienced advisors.
- High visibility in a fast-paced, execution-driven environment.
- Competitive pay, equity, and benefits.