PROJECT DESCRIPTION
Developed a RAG platform for structured and unstructured data, integrating ingestion, cleaning, chunking, hybrid FAISS/BM25 search, and language models; deployed using Docker and AWS EC2.
DATA FLOW
A view of the system.
- 01Source data
- 02Ingestion & cleaning
- 03Chunking
- 04FAISS / BM25
- 05Language models