Learn how to effectively implement domain-specific Large Language Models in this conference talk from Conf42 Prompt Engineering 2024. Explore the LEAPS framework and generative AI capabilities before diving into practical application architecture and key components. Master RAG-based inference engines and understand the importance of evaluation through metrics-driven development. Discover essential data preprocessing techniques, query enhancement strategies, and advanced retriever and re-ranking methods. Examine modular RAG architecture design and future implementation possibilities through real-world examples and best practices. The 22-minute presentation concludes with valuable insights and an interactive Q&A session to address specific implementation challenges.
Practical Strategies for Implementing Domain-Specific LLMs
Overview
Syllabus
Introduction and Speaker Background
Overview of LEAPS and Gen AI Abilities
App Architecture and Key Components
RAG Based Inference Engine
Evaluation and Metrics Driven Development
Data Preprocessing Techniques
Query Enhancement and Routing
Retriever and Re-ranking Techniques
Modular RAG Architecture and Future Steps
Conclusion and Q&A
Taught by
Conf42
