Lessons From A Year Building With LLMs

Learn from six leading practitioners who co-authored the influential O'Reilly article on Applied LLMs as they share hard-won insights from building production LLM systems at scale. This 35-minute conference keynote, recorded live at the AI Engineer World's Fair in San Francisco, presents a comprehensive three-tier framework for LLM implementation covering strategic, operational, and tactical considerations. Discover strategic approaches to LLM adoption and business integration from Bryan Bischof (Head of AI at Hex) and Charles Frye (AI Engineer at Modal Labs), explore operational best practices for production deployment and system architecture from Hamel Husain (independent AI consultant and former Google Brain researcher) and Jason Liu (AI consultant specializing in structured outputs and RAG), and master tactical implementation techniques from Eugene Yan (ML systems engineer) and Shreya Shankar (PhD candidate at UC Berkeley and ML engineering consultant). Gain actionable insights from practitioners who have successfully navigated the challenges of building real-world LLM applications, from initial strategy through operational deployment to tactical execution, drawing from their collective experience at organizations including Google Brain, Facebook, GitHub, Stitch Fix, and leading AI startups.