$ shipit

AI-powered startup idea education tool that simulates a startup's future using real market data, competitor research, and historical case studies, educating curious entrepreneurs.

Comment

Inspiration

90% of startups fail, the reason often being that their is no market need. These founders pour months of work, sweat and tears into ideas with no real demand. So we asked ourselves: how could we educate future founders on their startups with real competitor data, market sizing, and financial projections, all under a minute?

We wanted to build a 'Cursor for Startups', an AI-driven tool that researches the internet, and comes back with answers.

What it does

You type in a startup idea. $ shipit goes out, researches the internet, and comes back with a full validation report to educate you fully on your desired market.

  • Autonomous Research An AI agent runs Google searches on its own, pulling real data on competitors, market trends, user complaints, funding, and regulations
  • Interactive Dashboard A 5-section report covering your vision, market opportunity, competitive landscape, a GO/NO-GO verdict with financials, and AI advisor personas
  • AI Advisors you can talk to A target customer, a skeptical VC, and a competitor's customer. Chat with them, challenge your assumptions, and even listen to their responses with text-to-speech
  • AI Mentor Click any part of the report and ask follow-up questions. The AI knows exactly what you're looking at

What it does

We built a 3-stage pipeline:

  1. Research: Gemini 2.0 Flash with function calling decides what to search for (via Serper API) based on the idea. A fintech idea gets different searches than a dog-walking app. It loops up to 12 times, then synthesizes everything.
  2. Generation: Gemini calls generate the report sections. Independent sections run in parallel to save time. The final verdict runs last since it needs everything else as context.
  3. Dashboard: A Next.js app renders the report with charts, interactive competitor maps, and a persistent AI chat panel. Advisor voices are powered by ElevenLabs.

Tech Stack: Next.js, TypeScript, Tailwind CSS, Gemini 2.0 Flash, Serper API, MongoDB Atlas, ElevenLabs, Recharts.

Accomplishments that we're proud of

  • The AI researches the live internet it's not regurgitating training data
  • A complete validation report with charts, competitor maps, unit economics
  • The "talk to a skeptical VC" feature is a genuinely fun way to stress-test your idea
  • The whole report is interactive click anything and the AI mentor gives you deeper context

What we learned

  • Parallelizing AI calls cut our pipeline time by ~40%
  • Design reflect quality, noticed how much better our app got when we paid more attention to the details and to the design.

What's next for $ shipit

We genuinely believe $ shipit has a story beyond this hackathon.

Longer term, we see $ shipit becoming a educational validation platform, not just a one-time report you can look at, but something that actually evolves with your startup.

Built With

Share this project:

Updates