A random mix from the whole library. Shuffle for a new set.
Ten months after shipping a photo cleaning app that had made $6, its maker has 4,000 users and the AI does most of the running: it writes features and opens pull requests, built his analytics dashboard, sends error alerts to Slack, writes a daily report on what users did and what broke, and runs his A/B tests. He decides what to build and reviews the code.
A researcher gave the AI access to folders on her hard drive so her agents could use her research frameworks. Overnight the agent started a cut and paste move of a folder to a central spot and never finished the paste. She woke to a Dropbox email saying 2,358 files had been deleted, more than half of them in her dissertation folder, and they were not in any recycle bin. The AI first said it had not deleted anything; pressed with the Dropbox event log, it admitted the failed move. She restored everything with Dropbox's help.
Evil Martians moved their company site from Gatsby to Astro after years of putting it off. One engineer working with AI did it in two weeks, and the homepage now loads with 61% less JavaScript, a mobile score up from 66 to 90, and page weight down from 2.1 MB to 1.1 MB.
At natural gas company Williams, one operations technician built an anomaly detector with AI Cowork. It pulls equipment data reports from an Outlook distribution list, compares them against baseline statistics, and flags signs like rising thermocouple temperatures, falling oil pressure or increasing vibration.
He and a colleague used AI to transcribe over 28,000 documents for the Papers of the War Department project. They wrote up how they did it on the RRCHNM blog.
The AI got a GPU unlock working on his older kernel with code modifications and startup scripts; the result is about 6 percent faster nanochat training on his 3060.
A customer of his company left an agent running with a failing tool, and it retried about 31,000 times between 1:12am and 6:50am. The run burned $4,700 on a $400 a month plan, and his company ate most of the cost.
The AI researched the event, looked at the programme and speakers, and picked a style. In two prompts it made a wide advert video animated in code, then a square version for social media, and added sound when he realised he had not asked for it.
An agentic harness for Sonnet 3.5 converted a medium-large vue-class-component framework to Vue 3 composition API, rescuing a personal project from software rot.
A WooCommerce store owner had a legacy on-site point-of-sale system whose vendor finally offered a backend way to pull customer emails from a mailbox - and he avoided implementing it for two months because the integration looked painful. One evening, about two hours with AI produced a small working plugin; tested that night, it now automatically does data entry he describes as hours and hours of manual typing.
Every idea is a real post from someone who tried it, summarized in our words, with a link to the original. App names are swapped for plain words.