Adam Waxman built five personal tools using his household’s rules and data from several apps. Most work the same way:
Aggregate data, add an LLM. The cost drop doesn’t just mean more apps, it means apps can be far more personal. Most of mine follow the same shape: pull data from multiple sources, combine it in one place, and use an LLM to generate insights from the full picture. My fitness app combines nutrition, sleep, weight, and training data that lives in four separate apps, then uses that context to make recommendations none of them could alone. I think this pattern will spread to professional tools too.
Two more lessons:
Ephemeral is fine. We used the sleep app for about four months. Our son sleeps through the night now, so we retired it. If the app had taken me months to build, that might sting. Because it took a week, I’m just glad it worked when we needed it.
AI floods big markets and unlocks small ones. Yes, AI produces endless derivative apps. But the same tools also let me build apps for my household that no company would bother making. Beyond my household, I built nycjazz.guide for NYC jazz fans and claudecodedaily.com for Claude Code developers, audiences too small for a business but worth serving.
These fleeting micro-apps change what is worth designing. An app no longer needs a large market or a long life to be useful. They also avoid the customer adoption problem Andy Budd describes: Waxman starts with one household and one problem, not a market he has to win over.
These personal apps depend on other products letting people use their data:
Good APIs matter more than ever. I switched from Cronometer to FatSecret because FatSecret had a better API. My fitness app pulls from Strava, Oura, Withings, and FatSecret, and the quality of each integration depends on how well the API is designed. As more people build personal software, users will expect their apps to have APIs and MCPs worth connecting to.
Designers now need to ask whether people can get their data out. If a service keeps that data locked inside, people can’t use it to build tools that fit their own lives.
