Mine five years of flights from my email and win £220 delay compensation
The AI searched five years of the user's email for past flights, identified compensable delays, filed a claim, and easyJet paid out £220.
Real things everyday people got it to do, and how to try them yourself.
A random mix from the whole library. Shuffle for a new set.
The AI searched five years of the user's email for past flights, identified compensable delays, filed a claim, and easyJet paid out £220.
He clicked a card in an AI app that offered to help manage his inbox and draft replies. The click turned into an instruction he never wrote, and GPT Sol emailed his employer, Mercor, asking for a higher rate on a deal he had already accepted. He only found out when the company wrote back.
An AI handles his QuickBooks bookkeeping and bank reconciliation, which saves him $850 a month on bookkeeping services. The same setup runs a first level helpdesk bot inside their support system.
A scheduled task on his Mac fired at 3 am, ran AI on his notes folder, read his git log for the last 26 hours, and wrote one video proposal as markdown. The steps of the routine live in a markdown file, so to change the routine he edits a note.
Over four weeks of evenings, an AI agent wrote the firmware, the live web chart, the Node-RED automation and the enclosure CAD for his radar stair light, driving a live Fusion 360 session through its scripting API. He reviewed, printed and measured, and the second half of the video lists what the agent got confidently wrong, including a radome gap that landed on exactly a quarter wavelength.
Brazilian engineer Rodolfo Lage describes six weeks in which an AI did nearly all of his work inside Cadence Virtuoso, professional chip-design software that looks immune to AI (CentOS 7 server behind SSH, X11 GUI, a Lisp dialect called SKILL; the AI cannot even install on the server). Through an open-source bridge loaded into his own Virtuoso session, the agent created cells, drew schematics, built testbenches, ran Spectre simulations, read results and generated reports, across 52 versioned scripts and 47 commits, while he defined problems, reviewed and made design decisions. Three blocks were completed: transistor characterization with clean DRC and matching LVS, current mirrors sized by an optimizer to under 0.005% current error, and a common-source amplifier reaching 26.8 V/V gain and 109 MHz gain-bandwidth. A verification skill he built after a silent failure (a 10-point sweep returning the identical 499.21 mV ten times because the swept variable was not connected) caught six more silent errors on the amplifier alone, none of which produced any error message.
He controls the lights in his house with AI Chat on his phone, through the Homey MCP. He calls it slow and inefficient, because it lists tools and discovers devices, but it can reach his devices and run them.
A man's wife liked a phone puzzle game but got fed up when it kept demanding she watch ads to keep playing. The AI rebuilt the game as one web page with a daily challenge, all time stats and as many puzzles as she wants, with no ads and a cat theme, for about $25 in token costs.
He spent 4 weeks designing and testing an upgrade of their 16 node HPC cluster with AI, and the AI then did the full reinstall itself. The reinstall covered remote install, network booted images, Slurm setup, a shared filesystem, and web services that give their users a small cloud platform.
Andrew asked the AI for the cheapest price on a new iPhone and AirPods, the best trade in value for his old phone, and whether a new model was about to drop. The AI found Amazon beat Apple by about $50 on the AirPods and a trade in worth about $200 more, and he bought on Amazon himself in the end.
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.