← All projects
CASE STUDY
Domain-Agnostic RAG Chatbot
Retrieval-augmented chatbot with runtime provider switching
What it is
A flexible RAG platform that switches LLM & embedding providers at runtime — OpenAI, Anthropic, LM Studio, HuggingFace.
How it works
- LangChain pipeline over a ChromaDB vector store with sentence-transformers embeddings.
- Document ingestion for PDF / TXT / DOCX / Markdown with recursive chunking.
- Source-cited answers and real-time streaming over a FastAPI backend + Streamlit UI.
LangChainChromaDBsentence-transformersFastAPIStreamlitRAG