Most people use AI like a search engine instead of a teammate. After 30 years of coding, here’s how to delegate to AI like you would any developer and get brilliant results.
The story
Ever wonder why some people get AI to write brilliant code while others just get… meh?
After 30 years of coding and 8 years working with AI, I’ve figured out the difference isn’t the tool – it’s how you delegate to it.
Most people treat AI like a search engine. Type a question, get an answer, move on. But that’s like hiring a developer and only asking them yes-or-no questions.
Here’s what actually works:
Delegate everything at first. Yeah, even the stuff that feels inefficient. You’re not trying to save time yet – you’re mapping what your AI teammate can actually do. Where does it shine? Where does it fall flat? You need to know this.
Break things into chunks. Don’t hand AI a massive problem and hope for magic. Give clear steps. This isn’t just good for AI – it’s how you’d brief any developer. But with AI, it’s critical because you can isolate what went wrong and fix just that piece.
Iterate like crazy. First output is always a draft. Review it. Correct it. Feed those corrections back. The AI learns your style, your preferences, your standards. It gets better the more you work together.
Don’t delegate grunt work – delegate the automation of grunt work. Renaming variables? That’s for your IDE. But writing a script to automate tedious refactoring across your entire codebase? That’s perfect for AI.
The real skill isn’t prompting. It’s judgment – knowing what to hand off, what to keep, and how to orchestrate the collaboration.
Treat AI like a junior dev with weird superpowers, and suddenly you’re both unstoppable.
Recent Comments