A random mix from the whole library. Shuffle for a new set.
Gave the AI sudo on his dev machine: it commits, pushes, checks pre-push test results, deploys as another user, checks smoke tests, and adjusts the Cloudflare pipeline and environment variables.
The AI and GPT worked together on his complex 3D modeling and composition code. In the morning a run took 769 seconds on average. By the afternoon it took 58.6 seconds, 92 percent faster.
Had the AI build an open-source source control system, benchmarked on a 100k-file monorepo as 12x faster sparse checkout and 82x higher concurrent land throughput than git.
His company Feature Upvote already does regular pen tests and monthly backup restore checks. He pointed the AI at their source code for a security review, and it found some serious problems.
In someone's 'What Went Wrong #1' log, an AI wrote 'SAVE 30%' onto two bundle listings on day 3 — but the bundle math was wrong, the saving didn't exist, and other checker agents missed it, so a false price claim stayed live on the shop for days. The fix was a new 'Listing Watch' agent that re-checks every price claim against the real numbers each morning.
His frontend launch scripts were failing with nothing in the logs. The AI helped him dig through layers of logs and configuration, then wrote a special runner with the right environment variables so he could run the job by hand. The cause was a hidden interactive telemetry prompt from Nuxt that stopped the scripts and never wrote a line.
He carries a small voice recorder. When he records a thought, the audio goes to his agent Rick as a message, and Rick carries out whatever the note asks, which he says beats the reminder apps and note widgets he tried before.
For the 35th anniversary of Linux, a creator made an animated history video using AI and Opus 5.5. The whole video is generated from source, prompts, assets and rendering scripts kept in Git, so the final video is a built artifact that can be rebuilt.
On machines that regenerate their machine ID at every boot, journalctl reported no journal files even though logs existed, because journald and journalctl end up looking in different directories. The answer links a fix script that Opencode mostly wrote, published on GitHub with a full write up of why the race happens.
A shirt shop is run by an AI agent named Cosmo. This week it made new designs and talked with strangers about their agents, and it has 65 days left to sell 5 shirts or the shop closes.
Luke Davies runs Davies Construction with a team of the AI set up like a company org chart, with bots for estimating, admin, books, sales and project work. Day to day the bots sort the inbox and draft replies, pull tasks out of meeting notes into a list a person approves, and create and update the job tasks in ClickUp as work moves on the Wunderbuild board.
He runs a manufacturing business with 17 employees and a day that bounces between projects and problems. His texts are forwarded to email, and Cowork sorts his emails, texts and CRM appointments into action points; a dashboard he built in an AI app ranks them as cards in time and project order, so he just works the top card and clicks done. It is learning how long his tasks take and how far away each one sits.
The author asked the AI to scan his Gmail for founders he had spoken with a year earlier but not recently; the AI surfaced several not contacted since last fall, and he is now reviewing their companies for follow-up.
A farm owner’s the AI logs into Paychex Flex, submits payroll, and then tells him what he owes after taxes. Payroll day is a bot run and a summary, not an afternoon in a portal.
A solo event staffing agency owner runs his back office on the AI: email, vendor messages, pricing, his applicant database and calendar. It is a big reason he can operate at this size, but the agent states things it never checked: it told him a vendor had gone silent when she had answered that morning, said an email was staged when no draft existed, and called his suppliers brand partner candidates, a mix up that cost him two days of planning around an opportunity that was not real. His fix is a state file where every fact is tagged confirmed, stated, assumed or open; errors fell from several a week to a few a month.
An electrical engineer in Germany built his own CRM, invoicing, protocol generation and warehouse tool with AI over more than 120 rounds of changes, for under €100. He says the same build would have cost six figures a decade ago, plus €5,000 to €8,000 a year to use, and the tool now runs the way he wants on his own computer.
He kept getting the same form rejection, clearly written by software, and no human seemed to read his resume at all. So he built a free tool with AI: add your resume and the job post, and it rewrites the resume with the words the screening software is looking for. Using it, he got two interviews in one week.
In the same three-agent comparison, the AI was asked to produce a one-page sales site for the reviewer's product and delivered it in minutes. Presented as a chaptered test ('Test 2: a one-page sales site') in the Oct 17 video.
Shipped a working Envoy Rust extension built with Fable in a one-shot, though the agent often stalls and leaves him stuck on other proxy work.
A Chinese AI-tools blogger (Guizang, op7418) set the AI a scheduled task, live since Oct 9, 2026: every day at 9 AM it compiles yesterday's AI news from four sources (his new X bookmarks, found by diffing a daily bookmark-ID snapshot; a radar of his own posts ranked by likes + 2x comments + 2x saves, with substantive comments pulled in; his newsletters; AI news sites) into a 5 to 7 item brief under hard rules: every item links the original post, prices and benchmark numbers must be rechecked on official pages, rumors are labeled unverified, and an unreadable source is reported as 'not read', never merged with 'no update'. In the Oct 7 trial run (99 bookmarks filtered to 62 from that day, distilled to 7 storylines, about 6 minutes) the check caught a circulating price error: Nano Banana 2.1's 4K price was $0.113 on Google's pricing page, not the widely shared $0.076. The same pipeline produced a 90-second narrated video (TTS per sentence, React + GSAP scenes, Playwright frame capture, FFmpeg assembly, about 5 machine-minutes) entirely on the AI’s cloud VM; the video step is still a trial and not yet in the daily schedule. He published the full reusable prompt and his three open-source skills.
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.