When a single AI model is stuck or not providing the output you want, use this simple but effective technique. Make two different AI models compete against each other to produce a more accurate and refined result.
with Claire Vo
Start with a situation where an AI model (e.g., Claude) is giving you an incorrect or undesirable output despite several attempts to refine your prompt.
Collect three key pieces of information: 1. The original input or prompt you gave the first AI. 2. The unwanted output the first AI produced. 3. A clear example of the desired output you were trying to achieve.
Open a chat with a different AI model (e.g., Gemini). You will present your evidence to this second model to get a fresh perspective.
Present the context to the second AI. Explain that the first model failed and ask it to perform better by providing the correct output based on your examples.
Gemini, look at this garbage Claude gave me! My input was [Your Original Input], and it gave me [Unwanted AI Output]. The output I actually want is [Your Desired Output]. Surely you can do better. Please provide the correct response.
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.