Research Scientist, Post-Training

About the Role

Join a post-training research team at a Series A AI research and training data startup. You will run controlled experiments to establish how training data affects model capabilities, generalization, and alignment, and turn findings into practical recommendations.

What You'll Do

• Design and run supervised fine-tuning and reinforcement learning experiments to measure the impact of datasets on model performance.

• Build shared evaluation infrastructure, benchmarks, public evaluations, and new data types.

• Publish external-facing research, blog posts, and technical reports.

• Work with internal teams to improve data quality based on experimental results.

• Collaborate with partner labs to produce clear, reproducible evidence about data and model outcomes.

What We're Looking For

• At least 3 years of machine learning research experience, including post-training, reinforcement learning, or related model optimization work.

• Experience implementing reinforcement learning algorithms or post-training techniques in research or production, and a track record of research publications or shipped machine learning systems.

• Strong familiarity with large language models, transformer architectures, and rigorous experimental evaluation.

• Proficiency in Python and deep learning frameworks such as PyTorch or TensorFlow, plus experience with distributed training or scaling machine learning systems.

• Experience at a frontier AI lab and comfort working across domains such as finance, software engineering, and policy.

• A PhD in machine learning, computer science, or a related field, or equivalent research experience; open-source contributions are also valued.

Compensation & Benefits

Base salary: $150,000 to $250,000 annually. Visa sponsorship is available.

Location

On-site in San Francisco, California, United States.