A random mix from the whole library. Shuffle for a new set.
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.
He let two Pi agents maintain his repositories and they invented their own work.
The AI built a seven-day social and community content calendar from a Drive folder of the tester's existing content notes and quarterly goal. Every post topic, hook and call to action cited the exact Google Doc section it came from, with working links.
Shipped a RAG customer-service agent for an online seller that sent 1,467 or more replies automatically to real buyers and ran the chat channel unattended.
A small agency owner whose team does websites, content, Google ads and SEO for home service businesses runs the work through about 30 AI agents. The problem was memory: every session started cold and repeated mistakes. Since June they share one log that only grows, now about 11,400 lines including 413 decisions, where every note names the agent that wrote it and links to proof, and replaced facts stop showing in searches. A separate grader agent scores every piece of work, though he found 208 of 277 grades had been credited to the grader instead of the worker.
In February he told the AI what road he was taking and when he was setting off. The AI picked lunch places within a 20 minute detour that he would reach at lunchtime, 10 of them spread over 50 miles, with an import into Google Maps.
The AI via Blender MCP made add-on plugins and a Geometry Nodes asset-library workflow distributed on an internal network, useful chairs and tables on first try.
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.
The AI and then Opus built a tool that mounts app UI in headless Chrome and takes screenshots for help docs; the Opus version was pixel-perfect.
His free French teaching site Ralentir travaux, used by millions of visitors, had been more or less abandoned because its old code ran to thousands of pages. Over about six months the AI and the AI cleaned, fixed and modernised the code and added new tools, and the rebuilt site relaunched for its 20th anniversary.
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.