You will join the post-training research team at a Series A company that builds AI training data.
Your work is to design controlled experiments that show how training data shapes model capabilities, generalization, and alignment, and then turn those results into practical advice.
Responsibilities
Carry out controlled supervised fine-tuning (SFT) and reinforcement learning (RL) experiments that measure how a dataset changes model performance.
Help create public evaluations and new data types that improve how models are trained and assessed.
Build shared evaluation infrastructure and benchmarks so results can be reproduced across models, capabilities, and domains.
Write and publish external research, blog posts, and technical reports.
Work with internal teams to raise data quality using what the experiments show.
Requirements
0 to 4 years of experience in research or machine learning roles that build or deliver LLM training, evaluation, or post-training systems.
Experience at a frontier AI lab, and a record of designing, implementing, and analyzing controlled experiments on your own.
Solid understanding of LLM training and evaluation methods, including hands-on SFT and RL post-training.
Ability to design lightweight experiments and pull useful conclusions from noisy results.
Experience working across domains such as finance, software engineering, and policy.
Software engineering skills, including building or changing evaluation harnesses, training pipelines, benchmarks, or similar experimental infrastructure on your own.
Meaningful research or publications in model evaluation, post-training, or data curation.
Conditions
Base salary of $150,000 to $250,000 per year.
Visa sponsorship is available.
On-site work in San Francisco, California, United States.