Post-Training Applied Researcher

The role

You will post-train open-source models to outperform frontier closed models on specialized tasks. You will build reward functions, environments, eval harnesses, and training pipelines using complex, domain-specific datasets.

What you will do

• Design and run post-training pipelines including SFT, GRPO, DPO, RLVR, reward function engineering, and synthetic data generation.
• Build task-specific training environments and evals for domains like healthcare, code generation, and legal.
• Work directly with customers to translate production data into training signal.
• Run and analyze training experiments end-to-end.
• Publish findings at top venues and contribute to Baseten's open-source training libraries.

Required qualifications

• Hands-on experience training LLMs with reinforcement learning.
• Demonstrated understanding of GRPO or PPO including group advantage computation, clipped objectives, and KL penalty design.
• Strong intuition for reward engineering.
• Experience building multi-turn agent environments with tool use.
• Comfort working across the full pipeline from dataset construction through training, evaluation, and deployment.
• Experience with production ML systems.

Nice to have

• Preference for candidates who have closed a training–inference loop where production data feeds back into model improvement
• Experience with RL training frameworks
• Publications at NeurIPS, ICML, ICLR, focused on RL for LLMs, reward modeling, or alignment

Benefits

• Competitive compensation, including meaningful equity
• 100% coverage of medical, dental, and vision insurance for employee and dependents (U.S. only)
• Flexible PTO policy including company wide Winter Break
• Paid parental leave
• Fertility and family-building stipend through Carrot
• Company-facilitated 401(k) (U.S. only)
• Exposure to a variety of ML startups

About the company

Baseten powers mission-critical inference for AI companies by uniting applied AI research, flexible infrastructure, and seamless developer tooling. The company recently raised a $1.5B Series F.