Overview
The AI Document Q&A Assistant is a production-style Retrieval-Augmented Generation (RAG) system that answers natural-language questions over user-provided documents. It provides grounded, source-cited responses and supports multi-turn conversation. The system is built with LangChain, FAISS, OpenAI, FastAPI, and Streamlit, and is fully containerized and tested in CI. It ingests .txt and .pdf files, splits them into overlapping chunks, embeds them, and persists a FAISS index. For retrieval and generation, it embeds the incoming question, retrieves the top-k most similar chunks, and prompts the LLM to answer strictly from that context, citing sources. The API is a FastAPI service with request logging, session-scoped multi-turn memory, and typed request/response models. A Streamlit chat interface allows live demos and shows retrieved sources per answer. Tests include unit tests for chunking/loading and API contract tests, with all LLM/embedding calls mocked so they run free and fast.
Pricing
Free
Detailed plans have not been confirmed in our catalog. Check the official website for current limits and billing terms.
Visit WebsitePrices and limits may change. Confirm the currency, billing period, seat minimum and usage caps on the official website.
Use Cases
- Teams use the assistant to ask natural-language questions over their own documents and receive grounded, source-cited answers.
- Developers use the FastAPI service to integrate document Q&A into applications via a typed REST API.
- Users can engage in multi-turn conversations by passing a session_id to ask follow-up questions that resolve against prior turns.
- Teams can demo the assistant live using the Streamlit chat interface, which shows retrieved sources per answer.
- Developers can add their own .txt or .pdf documents to the data/sample_docs/ directory and rebuild the index.
- Engineers can run the system locally without Docker by following the quickstart steps.
- Teams can deploy the API and UI using Docker Compose for containerized operation.
Features
Answers questions about documents
Provides citations for answers
Remembers multi-turn conversations
Uses retrieval-augmented generation
Provides a FastAPI backend
Includes a Streamlit interface
Supports Dockerized deployment
Reviews
No reviews yet. Be the first to share your experience!