What to build

Ship a working AI marketing agent that connects founders (the clients) with influencers and creators, from start to finish. Every project needs these four parts:

1. Brief: the founder describes their product, audience, budget and goal (signups, sales, awareness), and the agent turns it into a structured campaign brief.
2. Match: the agent finds and ranks creators whose audience fits the brief, and explains each match (niche, audience, engagement, past content, price range). A list of the biggest accounts doesn't count as matching.
3. Outreach: the agent drafts a personalised pitch for each creator and handles the reply: interested, counter-offer or no.
4. Track: once a deal is agreed, the agent keeps track of deliverables, deadlines and results (clicks, signups or sales through tracked links or codes) and reports back to the founder.

Use any AI agent, any stack and any data source you have permission to use: platform APIs, creator databases, opt-in creator sign-ups or a seeded demo dataset. Some starting ideas:

• Two-sided marketplace: founders post briefs, creators opt in with a profile, and the agent proposes matches both sides can accept or pass on.
• Founder's copilot: a chat agent that takes "I'm launching a budgeting app for students" and comes back with a shortlist of creators, draft pitches and a suggested budget.
• Creator-side agent: an agent that works for creators, screening inbound brand deals and pitching itself to founders whose products fit.
• Deal tracker: the agent follows the campaign after launch, chases missed deliverables and shows which creators actually drove results.
• Onchain deals (optional): payment held in escrow and released when a post goes live and hits the agreed milestones.

Eligibility

• Solo builders and teams of up to 4.
• Anything started before the hackathon opened must be clearly marked, along with a description of the new work.
• You must build with at least one AI coding agent or assistant (Claude, Cursor, Codex, Copilot or similar).

Submission requirements

• A public GitHub repo with a README that explains how to run it, how the matching works, and which data sources and APIs it uses.
• A demo video (YouTube, Loom, Vimeo or X) of 3 minutes or less that shows the whole flow: a founder's brief, the ranked matches with reasons, a drafted pitch and the deal being tracked.
• A short description of what you built and why.
• A human approves outreach. The agent can draft messages, but a person has to approve them before they go to real creators. Demos should use test accounts, opt-in creators or seeded data.
• No API keys, tokens or scraped personal data in the repo. Use environment variables and include a .env.example file.
• One submission per person; you can update it until submissions close.

Judging

• Voting opens the moment submissions close and runs for 24 hours.
• Token holders vote, weighted by how much they hold at the snapshot (and still hold when they vote).
• The submission with the most vote weight wins. Ties go to the earlier submission.

Prize

• The winner receives the entire prize pool (sponsorship + 90% of creator fees earned so far).
• Every future fee claim is forwarded to the winner's payout address, forever.

Disqualification

Plagiarised work, non-functional demos or malicious code are removed from voting. So are:

• projects that send bulk unsolicited DMs or emails to real people;
• projects that scrape data in breach of a platform's terms;
• projects that fake followers, engagement or reviews;
• projects that leave out ad disclosures (like #ad) on sponsored content;
• any repo that exposes live API keys, credentials or personal data.