Weaviate Podcast Morningstar Intelligence Engine with Aravind Kesiraju - Weaviate Podcast #111!
Jan 8, 2025
Join Aravind Kesiraju, Principal Software Engineer at Morningstar, as he shares insights on the development of the Morningstar Intelligence Engine. They discuss the fascinating world of no-code/low-code AI applications and how to build advanced financial chatbots. Discover the intricacies of integrating diverse data sources and optimizing language models for financial tasks. Explore the evolution of Retrieval-Augmented Generation (RAG) data pipelines and the challenges of managing sensitive financial information while enhancing chatbot performance with intelligent question classification.
AI Snips
Chapters
Transcript
Episode notes
Morningstar Mo
- After GPT-3's release, Morningstar built 'Mo,' a GenAI application.
- Initial POC used FAISS, then Weaviate for its open-source nature and ease of use.
FAISS to Weaviate
- FAISS is suitable for in-memory vector search but lacks production-grade features.
- Building and scaling such systems requires significant investment, better spent on data ingestion.
RAG Pipeline Optimization
- Improve RAG pipelines with overlapping chunks, including paragraphs from next page.
- Reranking retrieved content after semantic search improves relevance and context for LLMs.