Baseten
Post-Training Applied Researcher
$200K–$275K
San Francisco · Hybrid · Full-time · Posted Oct 7, 2026
About the job
Post-train open-source models for customer-specific tasks by building training pipelines, reward functions, environments, and evaluations, then ship the models to production.
What you'll do
- Design and run post-training pipelines, including SFT, GRPO, DPO, RLVR, reward engineering, and synthetic data generation.
- Build customer-specific training environments and evaluations, and use production data to develop training signals and reward loops.
- Run and analyze end-to-end training experiments; publish findings and contribute to open-source training libraries.
What they ask for
- Hands-on LLM reinforcement learning experience, including demonstrated understanding of GRPO or PPO.
- Strong reward-engineering intuition.
- Experience building multi-turn agent environments with tool use.
- Experience across dataset construction, training, evaluation, and deployment; production ML systems experience.
Nice to have
- Experience with RL training frameworks.
- Publications at NeurIPS, ICML, or ICLR focused on RL for LLMs, reward modeling, or alignment.
Pay
$200K – $275K USD base salary
Tools
SFT, GRPO, DPO, RLVR, PPO
AI focus
Applied post-training research: fine-tuning and reinforcement learning for open-source LLMs, reward modeling, synthetic data generation, agent environments, and production feedback loops.
Model-written summary · the original posting is authoritative