Building with AI: the hard part isn't the AI
When you build with AI coding tools, the hardest part isn't the technology. It's being clear about what you actually want: who it's for, what could go wrong and what finished looks like. A vague request gets something that technically works but isn't quite right.
I've been working on an iPhone app this year with my co-founders at LIFEConnected AI.
The thing I didn't expect is that the hard part isn't the AI. It's being clear about what I actually want.
If I ask for something quick and vague, I get something that technically works but isn't quite right. When I slow down and explain who it's for, what could go wrong and what finished looks like, it gets a lot closer.
My background is business, not coding, so most of the build has been me directing AI coding tools and testing what comes back. Funny thing is, it's the same lesson I learnt running a business. Most problems come from a job that wasn't explained properly.
Anyone else using AI tools at work finding the same thing?
Co-Founder, Head of Product & AI at LIFEConnected AI. More about me