Worth Building

The piece argues that large language models have shifted the economics of building small tools: something that used to require justification and a path to monetization can now be created quickly and cheaply, so creators can prioritize craftsmanship over revenue. The author recounts building heade.rs, a domain lookup utility, by prompting Claude Code and deploying a working version to Cloudflare in about an hour; the tool shows DNS and registration records, live latency, email configuration and HTTPS details. The rest of the day went into polishing animations, graphics and interface; running costs are under a cup of coffee a month. Because the tool doesn’t have to pay for itself, it exists simply to be useful and pleasant to use.

That change lowers the bar for both programmers and non-programmers, enabling more personal projects, curiosities and niche utilities that serve a single user. It also creates a discoverability trade-off: cheaper software means more things shipped and more noise, so reaching a wider audience still requires giving people a reason to care through clear explanations of the problem and design choices. Some personal projects grow into larger products, but many don’t need to; making something that solves a small annoyance and delights its creator is often reason enough.