He sketched out how he wanted image collages to lay out and prompted the AI to build a simple tool with sliders for overlap and offset. He used it to make the image stacks for his project, kept improving it, and released Stacks V2 in October 2026 for anyone to try.
He built a virtual tabletop with AI, with proper dice, roll tables and boards, and he has been play-testing it a lot. He uses it when travelling and when his real table is buried, and he no longer needs Tabletop Simulator.
For the better part of a year an AI agent detected AI-related stories for his hcker.news feed and raised about 20 a day for his judgement, before he replaced the workflow with a fine-tuned classifier.
A developer had the AI build mazkir, a shared-memory service: tell the AI something once and it appears in his next chat, in an AI app and even in an AI app, stored as plain Markdown notes. Launch week brought about 160 visitors and exactly one sign-up — his friend.
Fulton's the AI flagged an Economist interview with Andrew Ng that matched his interests, wrote him a primer before it started, and placed it on both his personal and work calendars. He had not asked it to connect those dots.
A genealogist handed his AI an anonymized GEDCOM with about 3,000 individuals. The AI wrote a Python program drawing 300 frames into a 10-second MP4 animated pedigree (ancestors fade in by birth order, glow while alive, dim at death, living shown as 'Living'), then produced an 11x17 300-DPI print from a single frame.
A man had the AI help him write an image organizer and tagger in an afternoon. He is now running it on 20,000 meme photos he saved over the last 20 years. Each photo gets described and tagged for use in LightCraft, the photo app he is building.
His AI runs on a server with Ollama as the backend, so his data stays private. It keeps his todos in a simple text file, sends him reminders about upcoming due dates, and gives him an overview when he asks in a Telegram chat.
A father bought an old Pentium machine partly for retro games and found a childhood favourite, a 1991 learning game, froze the whole PC every time it loaded on the real hardware. He handed the AI the original game files. In five minutes it had taken the program apart, traced the hang to a sound check that waits forever on modern sound cards, and written a three byte patch into a copy of the file. The patched game worked on the first try, ready for his son.
His games crashed on the loading screen a few hours into a session. He pointed the AI at the problem, it suggested turning on the ntsync kernel module, applied the change to his NixOS setup and rebuilt it, and he played all day without a single crash.
He had his agent migrate his machine from Ubuntu to Omarchy with a backup taken first and a restore tested. The switch took about an hour, and he says he is not going back.
He tried many ways to show how certain artists became his favorites and how some faded out of his rotation. With the AI he ended up with a scrollable Sankey story of his listening, published on his dashboards site.
He set up a chain from his phone to the AI to his Windows PC and a background browser. The PC published a post to Bluesky and checked it had gone up, while his other screen work kept running separately.
After surfacing the Sept 28 triple-header, Mila built a sustainable weekly rhythm around Lixka's fixed classes, clinicals, and workouts, and asked permission before creating three tracked goals: quiz prep, reflection paper, and workouts/downtime.
A researcher wrote the Lean statement of a theorem himself, then fed the published paper to the AI and had it write the proof. It finished in a couple of hours, and the Lean kernel verified the proof.
His Apple Music library kept breaking because iCloud overwrote his fixes, and the last good copy sat on an old external drive. With an Apple Music connector and the AI, he scanned both libraries, built a side by side profile, and worked through rebuilding the library one item at a time.
This only partly worked. Worth knowing before you try it.
He has the AI run the upkeep on his main machine and his virtual servers: patches, cleanup, backups, and firewall checks. He reviewed about nine hours of its work sessions and says it may be the best systems administrator he has ever had.