Sequential QA Bot

Created a RAG architecture system to store and retrieve restaurant information, enabling intelligent routing between vector store knowledge bases and web search.

Integrated multi-agent sequential workflow using LangChain with conditional routing, LLM nodes, and hallucination detection.

RAG Architecture

• Vector store for restaurant information
• Embedding generation and similarity search
• Knowledge base retrieval
• Web search integration for real-time data

Intelligent Routing

• Multi-agent sequential workflow
• Conditional routing between knowledge bases
• LLM nodes for decision making
• Hallucination detection and prevention

Built with Python using LangChain framework for agent orchestration. Utilizes vector stores for semantic search and embeddings for knowledge representation. Implements retrieval-augmented generation (RAG) pattern combining vector databases with web search for accurate and up-to-date restaurant information retrieval.