A Retrieval-Augmented Generation (RAG) chatbot that combines semantic search with a large language model to deliver context-aware answers.
- Semantic search with embeddings
- Context-aware responses (RAG pipeline)
- Fast vector search using Pinecone
- Chat memory with conversation history
- Clean UI via Streamlit
- Configurable system prompts (YAML)
- GPU support (optional)
- Python
- Streamlit
- Pinecone
- SentenceTransformers
- Hugging Face Inference API
- PyTorch