This project is a modular LLM-powered service to answer questions based on documents and integrates flight number agent.
app.py: FastAPI app entrypointconfig.py: Configuration and secrets managementmodels.py: Request and response schema modelsauthorization.py: API token validationflight_api.py: Flight number API integrationtext_extraction.py: Document text extraction & chunkingretrieval.py: BM25 and FAISS retrieval logicembedding.py: Embedding model handlinganswer_generation.py: LLM answer generationutils.py: Helper functionstests/: Unit and integration tests
Install dependencies from requirements.txt and set environment variables for your API keys and secrets.
Run the app with: uvicorn app:app --host 0.0.0.0 --port 8000 --workers 1