Multi-user email RAG: E5-small-v2 + Mistral 7B GGUF over 10K+ emails, JWT auth, SQL-enforced per-user isolation (pgvector cosine), FastAPI backend.
-
Updated
Jul 14, 2026 - Python
Multi-user email RAG: E5-small-v2 + Mistral 7B GGUF over 10K+ emails, JWT auth, SQL-enforced per-user isolation (pgvector cosine), FastAPI backend.
661K-vector semantic indexing across 6 datasets. e5-large-v2 embeddings, FAISS IndexFlatIP. Foundational batch (superseded by batch-02).
Large-scale semantic indexing pipelines producing 8.35M+ vectors across Wikipedia, ArXiv, and StackExchange using e5-large-v2 and FAISS.
Semantic search framework for research archives. FAISS indices, e5-large-v2 embeddings, 4,600+ docs, 10 institutions, methodology docs.
Voice-enabled Hindi RAG over a 3.43M-vector Qdrant index. Speech streams to Sarvam realtime STT as you talk, queries are answered by hybrid dense + sparse BM25 retrieval with reciprocal rank fusion, and answers stream back token-by-token through four guardrail stages. FastAPI + React. HH Goa 2026.
To associate your repository with the e5-embeddings topic, visit your repo's landing page and select "manage topics."