The RL Fine-Tuning Playbook: CoreWeave's Kyle Corbitt on GRPO, Rubrics, Environments, Reward Hacking

About this episode

Kyle Corbitt discusses reinforcement learning and custom fine-tuning techniques for AI models.

Kyle Corbitt, founder of OpenPipe, breaks down reinforcement learning and custom fine-tuning for modern AI models. He explains how RL differs from supervised fine-tuning, why GRPO and LLM-as-judge post-training matter, and how these techniques can improve performance, latency, and cost on open source models. The conversation also covers reward hacking, evaluation design, LoRA adapters, and how Chinese labs are using distillation to fast-follow frontier models. Sponsors: Sequence: Sequence handles the full revenue workflow for complex pricing, from quoting and metering to invoicing, revenue recognition, and collections. Book a public demo at https://sequencehq.com and use code Cognizant in the source field to save 20% off year one AvePoint: AvePoint is building the control layer for AI agents so you can securely govern, audit, and recover every action at scale. Design trusted agentic outcomes from day one at https://avpt.co/tcr VCX: VCX, by Fundrise, is the public ticker for private tech, giving everyday investors access to high-growth private companies in AI, space, defense tech, and more. Learn how to invest at https://getvcx.com Claude: Claude by Anthropic is an AI collaborator…

People in this episode

Hosts: Erik Torenberg, Nathan Labenz

Guest: Kyle Corbitt

Topics covered

• reinforcement learning
• AI models
• fine-tuning
• reward hacking
• evaluation design
• open source models

Keywords

• reinforcement learning
• fine-tuning
• GRPO
• reward hacking
• evaluation design
• LoRA adapters
• distillation

Sponsors

Sequence, AvePoint, VCX, Claude

Mentioned in this episode

Organizations: OpenPipe, Fundrise

More episodes of "The Cognitive Revolution"

• AI:AM Highlights: Welcome to the AGI Era · September 5, 2026 · 2h 21m
• Write, Change, Recall, Forget: MongoDB's Pete Johnson on How Retrieval Drives Agent Performance · September 1, 2026 · 1h 37m
• AI:AM Highlights: Recursive Self-Improvement, Rushed and Vibe-Coded? · August 28, 2026 · 2h 12m
• RL's a Hell of a Drug: Metagaming, Reward Seeking & Motivated CoT Reasoning – Bronson Schoen, Apollo · August 26, 2026 · 2h 14m
• AI in the AM — Weekly Highlights: Relaunch Week (Aug 17–20, 2026) · August 22, 2026 · 2h 34m
• Let There Be Germicidal Light: This $500 Fixture Could Stop the Next Pandemic, from Complex Systems · August 16, 2026 · 1h 25m