Step 1 of 5 — Ingest Data
Upload Documents
Add hospital documents to the RAG vector store. Files are automatically chunked, embedded, and indexed into ChromaDB.
0
Documents
0
Chunks
0
Embedded
Processing Pipeline
01
Upload
File received by FastAPI
→
02
Chunk
Split into ~512 token chunks
→
03
Embed
Vectorize with OpenAI
→
04
Index
Store in ChromaDB