Sometimes, when developing an idea, especially a technical one, I use frameworks, like Edward de Bono’s Thinking Hats, the SCAMPER method or First Principles.
A few days ago I made a site which summarizes the steps involved in some common frameworks. As an experiment I implemented a cheap to run AI model which takes the summary of an idea you type in, and generates questions based on the steps of each framework.
After typing in some ideas I’ve been working on, I was surprised at the quality of questions I was getting back, and I even received positive feedback from a few people I sent it too!
So far, I’ve especially valued the questioning of assumptions which I might not have immediately thought of even with the frameworks in front of me!
“You could just use ChatGPT/LLM!”
Sure! and I’ve made a few experiments involving direct ideation with AI, including multi-agent setups which each answer for specific parts of thinking frameworks, however I often enjoy a more manual process, and I think this combination of a basic UI with integrated AI for generating questions, is a nice balance!
How does it work?
It’s almost as simple as it looks! The prompt is sent to the model augmented with the idea typed in and selected framework, and we get structured questions back which get placed into the UI and also cached in local storage in the browser — so once questions are generated, switching between frameworks is fast, and we don’t ask the AI for the same thing twice!
Is the AI costing you?
A small amount, but nothing to worry about (for now 😆)
Feedback
I’d love any feedback! Please email me!
