Presented at AWS re:Invent 2025, this talk dives into the unique challenges of fine-tuning LLMs for multi-agent systems. Learn about training data curation, reward modeling for agent collaboration, and deployment strategies for orchestrated AI agents at scale.