RAGxthon 2026 - Build the Future of AI Retrieval

About the Hackathon: RAGxthon 2026 is a virtual hackathon focused on Retrieval-Augmented Generation (RAG). Participants will design and build intelligent AI systems that combine Large Language Models with external knowledge sources to provide accurate and context-aware responses. The goal of this hackathon is to encourage developers, students, and AI enthusiasts to explore innovative ways to build AI applications powered by RAG pipelines, vector databases, and modern LLM frameworks. Participants will work individually or in teams to develop creative solutions such as AI assistants, knowledge retrieval systems, document Q&A tools, or intelligent search systems. Submission Requirements: Participants must submit: GitHub repository containing the project code A clear system architecture diagram showing the flow of the project Explanation of the RAG pipeline and components used with neat documentation Demo video or screenshots The architecture diagram should clearly illustrate the end-to-end flow, including: Data Source → Embeddings → Vector Database → Retrieval → LLM → Generated Response. Important: The system architecture diagram must clearly explain the flow of the project and how RAG components interact in the solution. Evaluation Criteria: Projects will be evaluated based on: Innovation & Creativity Quality of RAG Implementation Technical Execution Usability and Practical Impact Documentation and Presentation Rewards & Recognition: Participation Certificates for all participants Winner Recognition for top projects Dataset / Data Source: Participants may use open datasets, publicly available documents, APIs, or their own curated datasets relevant to the problem statement. The dataset should be suitable for Retrieval-Augmented Generation pipelines and must be clearly documented in the project submission. Recommended Technologies: Participants may use any technology stack. Suggested tools include: LLM APIs (Open-source or hosted models) Vector Databases Embedding models RAG frameworks Web frameworks for building interfaces Participants are free to innovate and choose their own implementation approach. Rules: All projects must be original work created during the hackathon. Participants may use open-source libraries and frameworks. Plagiarism or copied projects will lead to disqualification. Participants must submit their project before the submission deadline. The decision of the organizers will be final and binding.