Gita Gyan

Dharmic RAG

Ask a question, get an answer from the Bhagavad Gita โ€” chapter and verse cited, every time.

A retrieval-augmented system built on the Bhagavad Gita: it searches by meaning across all 700 verses and answers only from what's actually in the text, with the exact chapter and verse cited alongside the original Sanskrit.

FastAPILangChainChromaDBPython

0

Chapters indexed

0+

Verses embedded

Cited

Every answer sourced

เฅ

What it does

Verse-grounded answers

Every response is generated only from retrieved verses, never from the model's own memory of the text, and names the exact chapter and verse it drew from.

Semantic verse search

Ask in plain English or Hindi and retrieve the most relevant verses by meaning rather than keyword โ€” grief and เคฆเฅเคƒเค– surface the same passages.

Sanskrit alongside translation

Every cited verse shows the original Sanskrit next to an English translation, instead of paraphrasing it away.

Chapter-wise browsing

Explore the Gita chapter by chapter alongside the Q&A view, for reading start to finish rather than only querying.

How it's built

FastAPI backend
ChromaDB vector store
LangChain retrieval
LLM synthesis

Curious how Gita Gyan works under the hood?

Full source, architecture docs, and setup instructions are on GitHub.

View source on GitHub

This is a personal project and isn't publicly hosted, to keep rate limits and data resourcing manageable. Want to try it? Join the waitlist above or fork the repo and run it yourself.