With the start of baseball season, I revisited https://www.plai-ball.com/ for the new year. It was my first AI coding project, and it’s amazing how far things have come in a year.

The first version had a narrow AI integration: I wrote most of the code myself, using AI to learn specific tasks like how to structure API calls.

A year later, I’ve ceded much of that control. Outcome-based programming feels normal now: stating what I want and letting AI handle the rest.

This working model frees me up to focus more on the product. For example, I’m experimenting with two voices bantering back and forth to give the app more personality.

That product focus highlights how hard it is to go from prototype to polish. The code is the easy part now; the hard part is the hundreds of small tweaks that hone the product experience.

As for Plai Ball, the podcast audio still sounds mechanical. But I’ll keep iterating, refining, and having fun!

Originally posted on Bluesky by @monsur.hossa.in Source: https://bsky.app/profile/did:plc:je5uq7nujbrafiv4mcxcso6t/post/3ml5py63dlc2s