A random mix from the whole library. Shuffle for a new set.
He built a team of AI agents that run on a schedule while he sleeps. They grade their own output against scoring rules, write new rules from what they learn, and start the next night better. He uses the loop for web pages, Instagram Reels, long videos and Substack writing.
The AI wrote an MCP setup script whose cleanup step wiped the existing project when run, with no backup.
A man ran a local AI on his own machine that submitted almost 200 job applications over almost 5 days. His only part in the applications was clicking the captcha boxes, and running the work locally saved at least $400 in token costs at the AI rates.
He is job hunting in tech, aiming at startups and big tech. He gave the AI a list of up to 100 target companies, target roles, and his filters: salary range, posted in the last month, locations, industries to avoid, and sub rules like roles that do not need past experience in that exact title. Twice a week the AI sweeps the list and brings back matches. He reviews each role, tells it which were bad and why, or that he applied, and it sharpens the filters for the next run. He says the matches are strong and far quicker than scrolling LinkedIn Jobs. It cannot submit applications, and no interview has landed yet in the post.
LLM maintains the taxonomy and index for his automatically collected browsing archive in an Obsidian vault, used for his writing.
Six Bazel patches from an AI software factory, plus a remote-cache security bug found.
Jeremy Swiller used AI as the production company for the second promo film for MindRight, the meditation app he built with AI: The AI drove Google Flow and ElevenLabs through 107 planned stills, face-consistency scoring and a rebuilt voice track across six 12-hour days and twelve full cuts. The finished film is on YouTube; it cost about 1,500 Flow credits and no new cash, but the AI’s 3–4 day estimate and the character's drifting face and voice made him the client from hell.
The AI modernized a 22-year-old legacy system, updating dependencies, auditing security and adding features, raising test coverage from about 65 percent to about 80 percent.
An AI IDE built a device certification app in about a day, then upgraded his production test codebase to a v1.0-ready state with automated configs, error handling and simulators over two weeks.
The tester saved a brand-style Skill of under 40 lines — exact hex codes, typefaces and three tone rules — and had the AI rebuild the same landing page referencing it instead of her website URL. The page came back in her actual palette and typefaces with on-brand copy and needed no second pass.
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.