Adopt a structured, software engineering-inspired approach to iterating on AI designs. Use forking to safely explore new ideas and apply a two-step debugging prompt to reliably fix errors without frustrating back-and-forth.
with Claire Vo
When you have a working version of a prototype (a 'baseline'), create a duplicate or 'fork' before trying to add a new feature or make a change. This allows you to explore ideas in parallel without breaking your stable starting point.
Keep your project organized by clearly naming your forks. For example: Baseline - Profile Page, Var 1 - Add Edit Button, Var 2 - New Layout. This makes it easy to track different design directions.
When the AI doesn't follow an instruction correctly (e.g., fails to delete an element), first ask it to diagnose the problem *without* writing any code. This forces the AI to analyze the root cause.
Tried to delete this but it didn't work. Can you explain why? Don't write any code.After the AI has explained the 'why' behind the problem, you can then ask it to implement the fix. This 'explain, then solve' pattern leads to much more reliable and accurate results when debugging.
Follow Sahil Lavingia's method for delegating small but impactful maintenance tasks, like fixing an internal Slack report, to the AI agent Devin. This workflow shows how to use simple, clear prompts to fix code and update system prompts without distracting your engineering team.
Learn Gumroad CEO Sahil Lavingia's process for rapidly prototyping a UI with v0, then using the AI agent Devin to autonomously code, implement, and submit the feature for review, turning a multi-day task into a one-hour workflow.
Improve your writing in real-time by using ChatGPT as a first reader. Get instant feedback on structure, clarity, and flow as you write, allowing you to polish your work section by section.
We've made ChatPRD affordable so everyone from engineers to founders to Chief Product Officers can benefit from an AI PM.