Propose Birds of a Feather (BoF) sessions at IndiaFOSS 2024

BoF Sessions

Birds of a Feather sessions (or BoFs) at IndiaFOSS 2024 are informal gatherings for participants to discuss a particular topic without a pre-planned agenda. Themes for the same can be (but not limited to) FOSS Projects, FOSS in academia/govt/social sector, Open Hardware, Open Science, Open Data, Public Policy etc.


  • Duration: Each BoF Session will last for 60 mins.
  • Proposals: Propose a BoF session in this forum thread.
  • Deadline: The deadline for BoF proposals is July 31st,2024.
  • The proposer will serve as the moderator and initiate the session. Proposers who wish to participate in the discussion but not serve as a moderator should mention [Moderator Required] in the title of their proposal.
  • The moderator’s role is to ensure that the discussion is interactive and everyone in the audience gets a chance to share their thoughts while making sure the discussion stays on topic.

Unlike panels and talks, BoFs foster engaging conversations rather than formal presentations. Proposals should demonstrate a clear concept for the interactive nature of the BoF.

1 Like

In my personal experience 30 mins isn’t enough. 1 hour felt like the right amount of time in my personal experience at PyCon India in 2023. More than 1 hour isn’t necessary either because the individuals/group can take the conversation forward on the conference venue floor, outside the designated space for BoF sessions.



I would like to propose following topic for BoF session:

  • Title: Power up LLMs to be Factual Ninjas using Knowledge Graphs for RAGs [Moderator Required]

  • Abstract:
    In the era of big data, large language models (LLMs) are becoming increasingly important for tasks like question answering, document analysis, and chatbot development. However, traditional LLMs often struggle with factual accuracy, reasoning, and handling complex information.

This BoF session will introduce and discuss GraphRAG, a novel technique that leverages the power of knowledge graphs to enhance LLMs. We will explore the limitations of traditional LLMs and the vector-only approach of RAG, and how GraphRAG bridges this gap by equipping LLMs with a deeper understanding of the world through enterprise data. This technique boosts their accuracy, reasoning abilities, and handling of complex information.

The session aims to foster an engaging conversation around the potential of GraphRAG, its real-world applications, and its benefits over vector-only approaches.