Key Challenges in Building Enterprise LLM Applications - Large Language Models in the Enterprise

Explore the key challenges in developing enterprise-level applications using Large Language Models (LLMs) in this comprehensive conference talk. Delve into the complexities faced by product leaders and engineers when moving beyond demo applications to build custom LLM solutions on enterprise data. Examine issues such as misaligned behavior, context length limitations, cost considerations, latency problems, reproducibility concerns, and evaluation difficulties. Learn from the experiences of a Chief Data Scientist who has led teams in delivering mission-critical LLM applications, and discover strategies to overcome these obstacles. Gain valuable insights into the realities of implementing LLMs in business settings, dispelling common misconceptions and addressing the gap between expectations and practical implementation.

– Introduction
– Agenda
– Misaligned Behavior
– Context Length
– Cost
– Latency
– Reproducibility
– Evaluation
QnA