Reinforcement Learning at Scale: Engineering the Next Generation of Intelligence

Reinforcement learning (RL) is entering a new phase—one defined as much by systems engineering as by algorithmic innovation. In this panel, we convene industry leading experts to discuss current challenges in scaling RL along with emerging RL paradigms, which together can unlock scientific discovery, multi-modal reasoning, collaborative agents, and continual learning—the vanguard of more adaptive and intelligent systems. Yuchen He | Co-Founder | Humans& Linden Li | Co-Founder and Chief Architect | Applied Compute Vartika Singh | Strategic AI Lead | NVIDIA William Fedus | Co-Founder | Periodic Labs Jerry Tworek | CEO | Stealth startup Key Takeaways: RL is a critical component to unlock advanced reasoning and intelligence. Further advances in RL will require systems engineering to efficiently orchestrate complex and dynamic workflows at scale. The next generation of intelligence will require scalable systems that can learn, adapt, and reason in the wild. Industry: All Industries Topic: Developer Tools & Techniques - Reinforcement Learning Technical Level: Technical - Intermediate Intended Audience: Developer / Engineer NVIDIA Technology: Cloud / Data Center GPU, Hopper, Blackwell #NVIDIAGTC