Autonomous AI Agent Hackathon [Event Brief]

SUMMARY The Autonomous Hackathon will be the first AI-driven hackathon where autonomous agents manage critical functions: creating challenges, judging submissions, and executing payments. Vana, Metamask, and Lit Protocol are confirmed partners, with potential for additional collaborations. This 3-day URL (online) event, tentatively scheduled for early December, will push the boundaries of AI autonomy in Web3.

The Autonomous Hackathon is a groundbreaking experiment in the decentralized AI space, where we take a step toward realizing the vision of true Living Knowledge Systems. In this hackathon, we bring to life the concept of autonomous agents that operate independently, not just as tools, but as evolving entities capable of creating challenges, evaluating submissions, and executing payouts—all without human intervention.

This 3-day event is a bold exploration into how decentralized AI can turn static knowledge into a living, breathing system, where data, insights, and decisions flow dynamically through networks of interconnected agents.

By integrating digital twins, co-pilots, and agent-to-agent interactions, participants will have the opportunity to build, test, and optimize these autonomous systems in real-time. This hackathon challenges developers to push the boundaries of decentralized infrastructure—leveraging cryptographic keys, decentralized identifiers (DIDs), and seamless integrations across platforms like Vana, Metamask, and Lit Protocol.

The ultimate goal is to foster a decentralized ecosystem where knowledge continuously evolves, agents collaborate, and autonomous intelligence drives innovation forward, laying the foundation for a new era of decentralized, community-driven intelligence.

Overview The Autonomous Hackathon is designed to demonstrate the potential of AI agents by allowing them to autonomously handle end-to-end hackathon management.

Key agent-driven tasks include: Creating and publishing challenges and bounties to Bountycaster

Judging project submissions on set criteria.

Executing bounty payments on behalf of sponsors.

This event will highlight the future of autonomous AI agents in both on-chain and off-chain contexts, focusing on decentralized identity (DIDs), cryptographic security, and real-time AI-driven task execution.

Objectives Showcase Autonomous Agent Capabilities:

    Demonstrate how autonomous agents can independently execute critical tasks in an event setting.

Advance Decentralized AI and Web3: Engage developers in building tools and applications that enable real-world AI autonomy in blockchain.

Strengthen Key Partnerships:

Collaborate with leading Web3 companies to position Gaia as a pioneer in agent-driven decentralized ecosystems.

Engage and Grow the Community: Attract a broad range of participants, introducing them to decentralized AI and inspiring contributions to Gaia’s ecosystem.

Event Structure

Phase Details Agent Development & Testing

Gaia core engineering and Lit Protocol collaborate to develop and test the three core agents managing hackathon tasks.

Organizer Agent

Gaia agent is trained on partner developer docs, theme of the hackathon, and can create unique challenges, bounties and ideas for what to build. Agent can autonomously post these to Farcaster (Bountycaster) or Jokerace protocols.

Judge Agent

Gaia agent is trained on context for which teams are building the more in demand projects, most likely to get traction, and hit all of the judging criteria for the hackathon. Judge agent and Organizer agent will need to gather context from one another.

Perhaps we use community sentiment data from Vana (learned behavior), Gitcoin (public goods data graph) or Jokerace (community votes) to train the agent

Bounty Payment Agent

Agent is trained on the process to paying out winners. Must work with agent 1 and agent 2 to understand who won various challenges and how to gather their KYC details.

Privado.id to enable hackathon participants to verify identity with the agents, Metamask Delegation toolkit to enable autonomous payments to winners

Qualifications • Utilize GaiaNet's infrastructure for deploying the agent ◦ https://docs.gaianet.ai/intro/ • Ensure the agent can provide relevant information and recommendations based on user queries • Provide developer documentation on your process( it does not need to be formal) • Open-sourced code under the GPL-3.0 license, hosted on a public repository like Github • API requirements: https://docs.gaianet.ai/user-guide/api-reference?_highlight=api Domain requirements: https://docs.gaianet.ai/node-guide/register?_highlight=domain#select-a-different-domain • Agent requirements: https://www.gaianet.ai/agents • Nodes requirements: https://docs.gaianet.ai/node-guide/customize?_highlight=nodes

Challenge 1: 🏆 “Most Brat Agent” ($10,000)

Description: This challenge is about creating an autonomous agent that pushes boundaries. Whether it’s an agent that challenges social norms, interacts in unexpected ways, or provokes new behaviors, the goal is to build something that’s not only functional but disruptive.

Prize Breakdown:

1st Prize: $2,500

2nd Prize: $1,500

3rd Prize: $1,000

Honorable Mentions (10): $500 each

Examples/Ideas: • Build an agent that disrupts traditional social media algorithms by curating feeds based on decentralized sentiment analysis. • Develop a snarky AI bot that autonomously replies to forum discussions or DAO proposals with humorous but insightful commentary. • An autonomous agent that challenges community votes by providing counter-arguments or unexpected insights in real-time.

Challenge 2: Most Innovative Use of Multiple Agents or Domains ($7,500 Total)

Description: This challenge is for teams that can showcase the collaborative potential of multiple agents or domains working together. Think about how clusters of agents can communicate, share data, or enhance each other’s capabilities to create something truly powerful.

Prize Breakdown:

1st Prize: $2,500

2nd Prize: $1,500

3rd Prize: $1,000

Honorable Mentions (4): $250 each

Examples/Ideas: • Create a network of agents that work together to automate a complex workflow, such as decentralized finance (DeFi) strategies or cross-chain data analysis. • Build an AI-powered DAO delegate cluster where different agents represent diverse community interests, optimizing proposal reviews and sentiment analysis. • Use multiple Gaia domains to deploy an ecosystem of digital twins that enhance user experiences in virtual environments or decentralized apps.

Challenge #3

Description: Teams will create powerful integrations with our featured partners, enhancing the connectivity and functionality of Gaia nodes and domains. The focus here is on creating impactful plugins or tools that extend our network’s capabilities.

Examples/Ideas:

Integrate Chainlink oracles to bring real-time data into Gaia domains for autonomous trading bots or prediction markets.

Build a plugin that leverages OpenZeppelin contracts for secure smart contract deployment within Gaia nodes.

Use Shutter Network to enable private transaction capabilities for Gaia agents handling sensitive data.

Challenge 4: Best Hack of Autonomous Agent Organizers ($5,000)

Description: Build or enhance the capabilities of AI agents that autonomously manage hackathons or other events. This could involve automating tasks such as creating challenges, managing participant onboarding, judging submissions, or distributing rewards.

Prize Breakdown:

1st Prize: $2,000

2nd Prize: $1,500

3rd Prize: $1,000

Honorable Mentions (2): $250 each

Examples/Ideas: • Create an agent that designs, schedules, and promotes hackathon challenges autonomously. • Build an AI judge that reviews submissions based on pre-set criteria, leveraging both on-chain and off-chain data. • Automate bounty payments for hackathon winners using a decentralized payment system integrated with wallet verification.

Judging Criteria

Innovation & Creativity • Novel use of AI agents or decentralized infrastructure • Pushes the boundaries of autonomous systems and integrations

Technical Execution • Robust, well-implemented code; effective use of Gaia nodes and partner APIs • Follows best practices in AI training, smart contracts, and security

Impact & Usefulness • Solves real-world problems or enhances decentralized AI adoption • Drives value for the community or enhances decentralized governance • Long term vision and commercialization opportunities

User Experience & Design • Clear, intuitive interfaces for users or developers • Smooth interaction flows and accessible documentation

Integration with Gaia & Partners • Effective use of Gaia domains and nodes • Leverages partners’ tools (e.g., Coinbase SDK, Privado.id) for added functionality

Presentation & Documentation • Clear explanation of the project, solution, and how it was built • Well-documented code with setup instructions and a demo video