Raydar
Research Scientist
$150K–$250K
San Francisco, California · Full-time · Posted Oct 16, 2026
- Experience
- 0 to 4 years of relevant experience.
About the job
Conduct machine learning experiments on how training data shapes model behavior, then turn findings into recommendations, evaluations, and data-quality improvements.
What you'll do
- Run controlled fine-tuning and reinforcement learning experiments and assess their results.
- Help build public evaluations and new training data; share findings through technical content.
- Work with internal teams and external partners to improve data quality and present evidence-based results.
What they ask for
- Strong familiarity with large language model training and evaluation methods.
- Hands-on experience independently setting up, running, and interpreting controlled experiments.
- Ability to run quick, small-scale tests and draw useful conclusions from noisy results.
- Comfort working across domains such as finance, software engineering, and policy.
Nice to have
- Prior experience at a leading AI research lab.
- Preference for building and testing over theorizing.
Pay
USD 150,000 to 250,000 per year (base salary)
Tools
large language models, fine-tuning, reinforcement learning
AI focus
Conduct research experiments measuring how training data affects model performance and behavior; contribute to evaluations and training data.
Model-written summary · the original posting is authoritative