Daniel Han Explores Kernels, RL, and Agent Reward Hacking

Unsloth’s Daniel Han delivered an advanced talk on kernels, reinforcement learning, and reward hacking in agents. The session focused on low-level optimization, agent training mechanisms, and the risks of misaligned reward signals when building autonomous systems.

2026-07-11 ~ 2026-07-12 · 2 related posts

• Deep Dive into Kernels, RL, and Agent Reward Hacking — AI Engineer · 2026-07-11
• Daniel Han on Kernels, RL, and Reward Hacking — zacharylipton · 2026-07-12