Supply Chain RAG Embeddings

Fine-tuned all-MiniLM-L6-v2 for supply-chain RAG retrieval (historical precedents, export controls, India sourcing, mitigation QA pairs).

Usage

from sentence_transformers import SentenceTransformer
model = SentenceTransformer("mathurvarun84/supply-chain-embeddings")
q = model.encode("Red Sea shipping disruption semiconductor")

Local project

Set in .env:

EMBEDDING_MODEL_PATH=mathurvarun84/supply-chain-embeddings

Then rebuild ChromaDB:

python scripts/build_rag_collections.py --flush