Why Build Enterprise RAG with Postgres?

➡️ Lifetime access to ADVANCED-inference Repo (incl. future additions): https://trelis.com/ADVANCED-inference/ ➡️ Thumbnail made with this tutorial: https://youtu.be/ThKYjTdkyP8 OTHER TRELIS LINKS: ➡️ Trelis Newsletter: https://blog.Trelis.com ➡️ Other Products from Trelis: https://Trelis.com/ VIDEO LINKS: - Slides: https://docs.google.com/presentation/d/1OPYtx5aUwtwz5I-t-gAuLsUAcThly5jZyhj9qmzO1lg/edit?usp=sharing - BASIC-inference repo: https://github.com/trelisresearch/basic-inference - Nomic Embeddings: https://huggingface.co/nomic-ai/nomic-embed-text-v1.5 TIMESTAMPS: 00:00 - Introduction to RAG and Postgres implementation 00:32 - Background on using Postgres vs third-party services 01:22 - Overview of PG vector and PG best match 02:05 - Video structure outline 02:33 - Demo of RAG application 03:24 - Document upload and chunking process 03:53 - Data collation and vector/text search explanation 04:40 - Speed optimization using Cerebras/Groq 05:36 - Advantages of using Postgres 06:27 - Vector vs text search comparison 07:42 - Combining search methods 08:38 - Performance comparison data 09:56 - Vector search explanation 11:21 - Text search and BM25 explanation 14:38 - Postgres tools overview 17:42 - Database setup and configuration 20:29 - Basic RAG implementation 23:33 - Document embedding process 27:42 - Search implementation 31:40 - Advanced features: - 31:40 BM25 optimization - 33:20 Text processing improvements - 35:40 Asynchronous database calls - 37:15 Performance evaluation - 39:30 Command line interface setup 42:20 - Document chunking strategies 47:55 - Batch processing implementation 52:30 - Search functionality demo 56:40 - Final implementation review 01:17:48 - Conclusion and resources