When GenAI meets Ancient Wisdom
Session Description
In a world captivated by modern AI, what happens when cutting-edge technology is applied to one of the oldest epics known to humanity? This session explores the creation of the Mahabharata Chatbot—an intelligent assistant built using Retrieval-Augmented Generation (RAG) and powered by Neo4j knowledge graphs and Google Gemini. By weaving together ancient Indian wisdom with the capabilities of Generative AI, the chatbot enables users to ask context-aware, nuanced questions about characters, events, and philosophies from the Mahabharata. Discover how this project blends structured data, semantic search, and LLMs to bring mythological storytelling into the digital age—offering not just answers, but insight.
Key Takeaways
References
Session Categories
Which track are you applying for?
Speakers
Siddhant Agarwal
Reviews
This is definitely a novel idea and may be fun for people to hear about and see. But on the other hand, this is a fairly basic project these days. I think I've seen two or three talks on other RAG models about other contexts.
In my opinion, this would need to move from proof of concept into actually running somewhere in production for it to merit a slot.
Key takeaways aren't provided in the proposal, so it's not clear what the FOSS angle in this talk is. And just to be clear, Google Gemini isn't open-source. Currently, the proposal focuses on tool usage and connecting tools, neither of which is currently a novel problem in this space.
While the idea of a Mahabharata Chatbot was considered novel and fun, the reviewers felt that the project was a fairly basic use of modern tools like Retrieval-Augmented Generation (RAG). The concern raised by the reviewers were that the proposal focuses on tool usage rather than a substantial open-source contribution, and the FOSS angle was not clear.