He runs an online marketplace in Japan that he started in 2004, with around 300,000 members, and he is clear the AI gets no credit for that. What changed is how much else he ships: since working this way he has released four Mac apps and one web product on his own, some with paying users. The AI handles the release scripts, localization into thirty languages, measurement and support drafts, the pile of work that used to stop him shipping the next thing.
His weekly reporting agent builds the report and then reviews only the rows that do not match between systems, which is the part he used to check by hand. His own review is down to about 15 minutes a week and the setup has run for months.
After two days of failing to find a product studio app that fit, this footwear seller got a prompt workflow from Shopify's own assistant, then had the AI teach him VS Code with AI to generate store sections from reference images, which let him drop paid landing-page builders. He then built his own app that bulk-turns product photos into studio-quality images (4 images in 40 seconds) at about a third of the cost of the app he liked but found too expensive.
Consultants at the same co-op connect the AI to their calendar, GitHub and ticket tool. The AI works out how their time split across projects, fills in the timesheet day by day with a reason for each day, and each person checks it and approves it before it is saved in Salesforce.
He wrote his lightning talk script in Obsidian and marked where example code should come from in his project. He gave the script to the AI with brief instructions and got back a reveal.js deck that was 95% done.
This only partly worked. Worth knowing before you try it.
The AI wrote a POSIX vi implementation from the POSIX specs over a weekend; he has used it for over a month with no fault found in its POSIX implementation.
After hackers hit their research site meta-analysis.cz, which had no central backup, the AI and the AI rebuilt it from Wayback Machine copies and scattered files on their hard drives. The site is now on GitHub, its datasets are archived on Zenodo with a DOI, and every paper has an HTML version next to the PDF.
Locally hosted agents rebuilt a Conductor flaw in an isolated lab and got root access inside the test container. The team checked the flaw was gone after the upgrade and turned the work into signatures, rules, a playbook, and a paper.