LLM Avalanche: Jerry Liu · Data Considerations for building more Production-ready LLM Applications

ai.bythebay.io Nov 2025, Oakland, full-stack AI conference Speaker: Jerry Liu, Co-founder/CEO at LlamaIndex Abstract: Large Language Models (LLM’s) are revolutionizing how users can search for, interact with, and generate new content. Some recent stacks around Retrieval Augmented Generation (RAG) have emerged where users are building apps such as chatbots/QA systems/agents on their own private data. These stacks are super easy to setup but very basic (text splitting, throw into a vector database, top-k retrieval) and may not be suitable for production. In this talk we talk about practical considerations you must take into account over your data to make your LLM app more performant/reliable in a production setting.