WatDo

• 

  Example showing graph flow.

Inspiration

The goal was to create a tool that helps users plan their next steps efficiently, regardless of their context. The idea was to provide structured yet diverse solutions to guide decision-making.

A close example is the state of computer science undergraduate students. Many students are ambitious and want to know where they are going. They can plot a path with watdo.me to get where they want to go to, or alternatively, work backwards from where they are with AI predictions.

What it does

WatDo generates three types of next steps for users based on their current situation:

Historical Approach – A decision based on past data and proven methods, meaning that accuracy is prioritized.

Progressive Approach – A forward-thinking, innovative decision, meaning that profitable and/or progressive decisions are prioritized.

Chaotic Approach – A random, out-of-the-box decision that's rare to occur, but something that's not impossible.

How I built it

Developed with React and React Flow to visualize logic paths. Prompts made via Cohere's Command+A API.

Challenges I ran into

Slow request handling, impacting node branching performance.

Accomplishments that I'm proud of

Successfully implementing React Flow to create interactive and logical pathways, AI tools like Cursor and v0 to help build the app.

What I learned

Plans always change, for the better more than not.

What's next for WatDo

Finer-tuned prompting; avoid outputs that don't help. Create an entire graph from two nodes - start and end, and give three milestones they must reach in that process. User can play around with nodes after. This can be a new feature or override.

Built With

• cursor
• react
• typescript