Unlock the Next Frontier of Intelligent Systems With Knowledge Graphs + AI

Transform the way we understand and interact with data by bridging the gap between knowledge graphs and artificial intelligence. The Hypermode Knowledge Graph + AI Challenge is your opportunity to revolutionize how intelligent systems process, connect, and derive insights.

Why Knowledge Graphs + AI

GenAI models still struggle with domain-specific knowledge. Traditional RAG partially solves the problem, but human language is not deterministic. As such, many AI apps fall well short of their promise and hallucinate frequently.

Knowledge graphs and GraphRAG are quickly becoming the best-in-class practice for creating shared understanding between humans and AI.

Tools like Neo4j and Dgraph provide a semantic layer that allows AI systems to understand relationships, context, and nuance in ways traditional databases cannot enable. This unlocks agentic workflows, where your agent has a dramatically better sense of what models, tools, and data to call because of it understands your data relationships.

Knowledge graphs provide the missing link - a powerful framework that maps out intricate relationships between data points, enabling AI to:

  • Enhance Contextual Intelligence: Move beyond surface-level pattern recognition to deep, nuanced understanding
  • Improve Reasoning Capabilities: Create AI systems that can navigate complex logical connections
  • Provide More Transparent Decision-Making: Trace the reasoning path of AI models with clear, interpretable connections

Modus: The Open Source Framework For Building Model-Native Apps With Knowledge Graphs

The Modus framework now enables seamless interaction between AI models and Dgraph and/or Neo4j knowledge graphs, allowing developers to:

  • Build context-aware AI applications with rich semantic understanding
  • Implement advanced GraphRAG patterns
  • Create more intelligent, relationship-driven AI agents
  • Leverage the full power of graph-based data representation

The Challenge: Innovate at the Intersection of Data and Intelligence

We're calling on developers, data scientists, researchers, and innovators to demonstrate breakthrough applications that leverage knowledge graphs to supercharge AI performance across industries.

Potential Focus Areas
  • Healthcare: Connecting medical research, patient data, and treatment outcomes
  • Financial Services: Mapping complex economic networks and risk analysis
  • Scientific Research: Accelerating discovery by connecting disparate research insights
  • Enterprise Knowledge Management: Creating intelligent, interconnected information systems
What You'll Get
  • $10k USD total prize pool
  • Exposure to cutting-edge AI and knowledge graph technologies
  • Mentorship from leading experts in the field
  • Potential collaboration opportunities with Hypermode
Who Should Participate
  • AI Engineers wanting to leverage knowledge graphs in their AI apps
  • Product engineers looking to add AI-backed features to their apps
  • Data scientists, machine learning engineers, research academics, software developers, and AI enthusiasts

Project Requirements

Our goal for this hackathon is to encourage creativity in the way that knowledge graphs can be used with AI models so we're leaving the guidelines as open ended as possible. However, when planning your project please consider the requirements below and also the prize categories for some hints at what the judges will be looking for.

Your project must:

  • Use the open source Modus API framework
  • Make use of at least one AI model via Modus
  • Make use of a knowledge graph, such as Dgraph or Neo4j (although we leave the specific knowledge graph implementation up to you)

Your Submission

  • Should be open source with the code publicly available on GitHub, be sure to include a link to the public GitHub repository in your submission.
  • Should clearly describe the goal(s) and implementation of your project, including how you used the Modus framework, which AI model(s) you used and the purpose, and describe the knowledge graph (Dgraph, Neo4j, etc.) used in your project. Include this information either in the README for your project on GitHub or in a blog post.
  • Include a short video demo of your project (optional).
  • Include a link to your deployed project so the judges can try it out (optional).

Examples of projects could include (but are not limited, feel free to use your imagination!):

  • A full stack web application that allows users to search for travel destinations, with the results tailored to their interests using an AI model and a knowledge graph alongside tool use / function calling to retrieve updated information
  • A GraphQL API built using Modus and deployed on Hypermode that implements an agentic workflow in response to a natural language query where an LLM model decides which data sources to use based on the users request.
  • Feel free to share your idea in the Hypermode Discord or in the Devpost message forum and we'll be happy to give you feedback on your idea!

1 winner

$3000 USD, paid via PayPal or ACH

1 winner

$2000 USD, paid via PayPal or ACH

1 winner

$1000 USD, paid via PayPal or ACH

1 winner

$1000 USD, paid via PayPal or ACH

1 winner

$1000 USD, paid via PayPal or ACH

1 winner

$1000 USD, paid via PayPal or ACH

Best Use of Knowledge Graph RAG

1 winner

$1000 USD, paid via PayPal or ACH

Submitting to this hackathon could earn you:

Raphael Derbier

Raphael Derbier
Staff Architect, Hypermode

Michael Hunger

Michael Hunger
Head of Product Innovation, Neo4j

William Lyon

William Lyon
Developer Experience, Hypermode

Derek Briggs

Derek Briggs
Head of Design, Hypermode

Jai Radhakrishnan

Jai Radhakrishnan
Software Engineer, Hypermode

Himanshu Sinha

Himanshu Sinha
Director of Advanced Data Science, Marriott International

Li (Colin) Qian

Li (Colin) Qian
Software Engineering Manager, Disney Streaming

Ryan Fox-Tyler

Ryan Fox-Tyler
SVP of Products and Engineering, Hypermode

Kevin Van Gundy

Kevin Van Gundy
CEO, Hypermode

Jessica Feng

Jessica Feng
CMO, Hypermode

  • Innovation
    The project's originality and creativity.
  • Relevance
    How well the project uses Modus, AI models, and knowledge graphs.
  • Presentation
    The quality and clarity of the demo and writeup, with videos encouraged but optional.
  • Educational value
    How effectively the project helps others understand the concepts it covers.
  • Completeness
    How much the project functions as a useful prototype.