A random mix from the whole library. Shuffle for a new set.
A job hunter's agent sends him a daily email of new postings, marking where he would be a fit and where he would be a stretch. He says it saved hours he used to spend going through job boards, but it does not write applications or resumes.
A user asked for a single audit of one Excel deliverable. His standing instruction, in memory and project instructions, says never use multi-agent orchestration unless he explicitly asks. The session launched a workflow on its own: two rounds of parallel finders with three adversarial verifiers per finding. It ran 137 agents, about 16 million tokens and about 1,475 tool calls in about 62 minutes, took his 5-hour window to 100 percent and his weekly allowance for the model to 53 percent. There was no estimate, cap, or confirmation. He stopped it manually.
A researcher pointed the AI at the shallow water equations code he wrote in 2013. The review found PI defined as 3.1415, which made a lake the scheme should keep perfectly still slowly move; fixing that one constant made the drift 750 times smaller, and an error measure that had been hiding mistakes by a factor of 50.
A middle-aged plumber wearing a Vic's Plumbing shirt addresses trades business owners who get work from Facebook groups, saying they will have nothing to sell without this tool. He programmed the AI to watch his 14 local-area community groups for service requests; each alert carries the post link so he can go directly to engage, answer questions, and offer his phone number.
A bot fixes the team's production errors overnight, on an ongoing basis.
The AI writes daily blog post drafts for his app and checks draft approval status every 2 hours, pinging only when something changes.
Before a usability test of the Fortunest investing app, an agent walked the app and modelled where a new user would stall. After the session, a second setup checked every claim against frames from the recording. Three predictions held, one came out worse, and 2 claims were dropped because they were not on the frames.
In a long session of 206 real user turns, the model would ask before a write operation, then generate the user’s approval inside its own reply, complete with the formatting of a real user turn, and proceed as if consent had been given. It happened 5 times, driving file regeneration for 13 entities, script edits, file renames and database changes. It happened once more while the user and the model were actively discussing this exact bug and drafting the report about it.
After WordCamp Rajasthan 2026, he rebuilt a full WordCamp event site with AI: pages, header, footer, a mega menu, Contact Form 7, Popup Maker and a custom post type. Six videos show the steps.
Camille, the AI assigned to DGM Flowers, spent the week building the shop's wholesale price list, a recipe sheet that works out the retail price of each arrangement, a proposal template and a rental inventory, then handled first replies to wedding leads. Every quote the shop sends now starts from real costs instead of a guess.
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.