Finding an idea
You don't need an idea before the hackathon. If you're stuck, start by looking for problems rather than technologies.
YC's Requests for Startups is a good place to see what problems founders and investors are interested in.
https://www.ycombinator.com/rfs
YC Library has a lot of practical advice on finding ideas, talking to users, and building startups.
https://www.ycombinator.com/library
You can also look at:
-
Hacker News: https://news.ycombinator.com/
-
Product Hunt: https://www.producthunt.com/
-
Google Trends: https://trends.google.com/
A useful question to ask is: "What do people currently do to deal with this problem?"
If the answer involves spreadsheets, copying and pasting, sending emails, manually checking websites, or using several different tools together, there might be something worth building.
Building with AI
AI coding tools can help you build much faster, but they can also make it easy to generate a lot of code without understanding what you built.
A good workflow is:
-
Build one small feature.
-
Test it.
-
Make sure you understand what the code is doing.
-
Fix anything that breaks.
-
Move on to the next feature.
Don't ask an AI tool to build your entire application at once and assume it works. A small project that works well is much better than a huge project that barely works.
Product resources
Before you start building, try to answer three questions:
-
Who is this for?
-
What problem do they have?
-
Why is your solution better than what they do now?
You don't need to build a startup in one day. You need to build enough of a product to prove that the idea works.
Technical resources
GitHub: https://github.com/
Vercel: https://vercel.com/
Supabase: https://supabase.com/
OpenAI: https://platform.openai.com/
Anthropic: https://console.anthropic.com/
Google AI: https://ai.google.dev/
Check each service's documentation for current API limits and pricing before using it.