May 27 · 15 min read · Series intro. I'm a non-CS solo dev who built and shipped a production stock screener almost entirely by "vibe coding" with an AI agent. The site works. Users use it. And it has cost me, in real ways,
Join discussion
May 18 · 6 min read · We made it easier to use. Then it broke. I got pulled into an incident recently where one of our highest-value enterprise accounts, couldn't export their survey data. Their analytics pipeline had gone
YJacob commentedMay 10 · 2 min read · I've run over 100 post-mortems. The worst ones end with 'Alice will be more careful.' The best ones end with 'we fixed the system.' Here's how you get from the first to the second. The language rule Ban these phrases from retros: 'Should have...' 'A...
Join discussion
Apr 25 · 6 min read · Yesterday I posted about losing $540 a month to silent user churn. The day after, I went looking at why my LinkedIn scraper itself was bleeding compute every run. Found three bugs that were quietly eating margin. Shipped fixes today. If you missed it...
Join discussionApr 25 · 7 min read · Anthropic's April 23 Postmortem: 3 Confounding Changes Behind Claude Code's Month-Long Quality Drop Like reading an aviation accident report. Engine fault. Pilot error. Weather. Each alone wouldn't have brought the plane down. Three together did. F...
Join discussionApr 24 · 7 min read · My Apify portfolio went from $268 rolling 30 day profit to $50 over about three weeks in April, and I spent most of that stretch assuming it was a reporting glitch. It wasn't a glitch. Two agency buyers who'd been running my Google Maps Lead Intel ac...
Join discussionApr 15 · 3 min read · The Post-Mortem Nobody Learns From I've sat through hundreds of post-mortems. Most follow the same pattern: something breaks, someone writes a Google Doc, we have a meeting, we list action items, nobody follows up, the same thing happens again in 3 m...
Join discussion
