Contact person: Web RAG Chatbot

Phone:Show

Email:Show

Location: Bangalore, India

Budget: Recommended by industry experts

Time to start: As soon as possible

Project description:
"This project centres on building a production-ready, retrieval-augmented AI chatbot that runs in the browser. I already have the content that the assistant must draw from, so the work is to create the full RAG pipeline—data ingestion, vector embedding, retrieval, and answer generation—then wrap it in a clean web interface.

Key elements of the build
• Retrieval-Augmented Generation: split, embed, and store my documents in a vector database (FAISS, Pinecone or similar).
• LLM orchestration: LangChain or an equivalent framework to combine retrieval with response generation through an API such as OpenAI or Anthropic.
• Web front end: a responsive chat UI with streaming responses and citation links back to source passages.
• Deployment: containerised code (Docker or similar) that can be launched on a standard cloud instance; environment variables should hold all secrets.
• Logging & monitoring hooks so I can track usage and spot failed calls.

Acceptance criteria
1. A user can ask any question on the site and receive an answer grounded in the supplied documents, with inline citations.
2. Answers degrade gracefully—if the corpus lacks information, the bot replies with a helpful fallback rather than hallucinating.
3. The full codebase, [login to view URL] / [login to view URL], and a concise README are delivered and run locally with a single command.
4. A short loom-style demo shows cold-start deployment and at least three sample queries.

Future phases may add mobile or social-media endpoints, so a modular design is appreciated, but for this milestone the deliverable is the web version only." (client-provided description)