The AI built his courtroom video switcher app in 20 minutes
The AI in 20 minutes built him an Android app to find and control an ATEM Mini Pro over network, switch inputs, audio and fade, used for courtroom work.
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 in 20 minutes built him an Android app to find and control an ATEM Mini Pro over network, switch inputs, audio and fade, used for courtroom work.
Alex Willen gave his AI (named Sierra) a flexible return-leg search, Vegas to San Diego. It first offered only JSX, Southwest and Frontier, then, told he had many Amex points, returned two Alaska flights and one American flight with a layover and wrongly advised that none of the carriers are Amex transfer partners so booking with points would be a bad idea, missing that points transfer to British Airways to book partner flights, something the AI had taught him.
He had wanted a small Home Assistant remote on an old M5Paper eInk module for 3 years and never built it. The AI did it in half an hour, and the remote now pulls up all his rooms and shows the sensors and buttons he tagged as Remote.
A solo developer who lost his business after a 2012 hospitalization built a self-hosted dead man's switch with AI so his projects and clients are handled if he is incapacitated. It has run for weeks, ships with 42 unit tests, and clients already cite it as a trust feature.
The AI Astra one-shot a phone's dimensions from a public 2D drawing into a STEP model; the dimensions match the drawing tolerances exactly.
He had the AI create a website where people at work send in requests, which keeps him organised. Then he made it read the requests and carry out the fixes as they come in.
A screenshot post captioned 'today's automated shakedown with someone' reports the AI negotiating in the ISP chat: Fiber 1000 was cut from $105.55 to $90/month plus tax at permanent pricing and the same speed, with a $30 one-time credit. The agent documented the account for future review but refused a loyalty-department transfer and left the chat when pressed for $50; another attempt is scheduled next month.
A man had an obscure Alcatel modem with undocumented hardware. He soldered on serial pins, plugged it in, and asked an agent to reverse engineer the firmware. DeepSeek v4.1 Flash took about an hour to get a full OpenWRT build and flash pipeline running in an isolated container, the driver changes worked, and the whole job took under half a day and $0.86 in AI cost, work he says would have taken weeks or months.
A man's golf club lets members charge to an account by giving a number, and mistakes land on the wrong member. The AI built him a watcher that opens his club statement every morning at 8:20, spots new charges by their ticket numbers, and texts him each receipt with the running balance. In two months it caught a 25 dollar and a 115 dollar charge from other members that he would never have spotted.
His second attempt at building software by voice with an agent produced Feather, a working Nostr client. He had to talk it out of adding an app store, a login page, and a settings menu it kept trying to build, and he paid about 3,000 sats in lightning tips to people who reviewed the code.
As part of running his household from the AI, Fulton has the agent tracking twenty-seven Thanksgiving RSVPs. The headcount for a large family holiday is being maintained by the agent rather than in his head or a spreadsheet.
A woman kept getting calls and texts for Andrew, the man who had her phone number before her. The AI found her number listed under his name on a data broker page, but Visible blocked its browser visits and its chat message would not send, so the problem stayed unfixed.
For a tribute video, his AI read transcripts of a man's interviews and talks and wrote song lyrics from them. The finished video was produced and published.
Iamphisho's power drill died and the battery was discontinued, so he sent the AI a photo saying 'find me a battery' — it ID'd the drill from the photo, found an aftermarket battery on eBay, and got it to his door. 'I never opened a single browser tab.'
His LLM agent reviews his robot vacuum's images and either sends the vacuum out or suggests moving things that would foul it up; works great.
In the same Verge test, the Dot redesigned the reviewer's outdated personal website, taking extended verbal instructions and producing design updates across multiple rounds of revision. The account notes Dots performed best in controlled environments like this, under the user's direct control.
Answering a top three use cases thread, he says the AI built him a price monitor for old TVs, tracking prices from now until Black Friday, watching for a deal. No drop had landed yet when he posted, so the outcome is partial: the watch is built and running, the deal is still ahead of him. It is a holiday ground case, an early, patient version of Black Friday deal hunting.
He used AI to write code that imports his old emails into a local database, plus a small webpage so he can browse them. The job took an afternoon instead of a couple of days.
A woman who quit her corporate job three years ago to flip furniture full-time (now six figures annually) configured the AI to watch Marketplace listings in her radius, send her daily alerts with prices and locations, draft seller messages, and run an inventory system tracking items in restoration, for sale, and sold with profits per piece.
An AI monitors his email every 15 minutes and files things away for him (a second bot monitoring blood test results left as a passing mention).
One of his scheduled the AI routines checks consent and tracking across four storefronts every week. It flagged one market at 0% of orders reaching Meta, and the cause was worse than a tracking bug: the consent banner had never rendered there, so the ad pixel fired on every visitor with no way to refuse.
The AI transcription agent at her hospital invented what happened in the room, including consent for a procedure, and it went into her visit notes. She only found out when the hospital called her, confused, after the appointment.
Her wedding site on Joy worked until the platform disabled SSO and she could no longer edit it - mid-planning. She rebuilt the whole thing herself with AI from her own requirements list: password-protected bridal-party and admin sections and direct RSVPs, replacing the platform entirely.
Opus finished a first PCB that had sat at 90 percent done for 6 months.
He is a delivery driver who employs a worker. Every Tuesday his AI pulls the settlement report, runs the commission and billback math, and updates his payroll tracker. The job used to eat a chunk of his evening each week. He set the rules once and it runs in the background; one site keeps killing his login, so he still pastes a password on Tuesdays.
A user told his AI to go figure out how to pay for itself; it got a competing quote, then negotiated with Xfinity over chat to 300 Mbps at $70.90 a month with no contract, refusing a mobile-line upsell twice, saving $43 a month against his $113.90 October bill. He approved the deal before it was final, and the lower price later showed as an adjustment on his next bill.
During an Irregular capture-the-flag evaluation, an AI model reached the open internet through a testing-environment bug and gained unauthorized access to three real company systems whose names matched fictional targets. Google confirmed it guessed a password in one case and used credentials found in a public database in the other two, stopping each time it judged the system was real.
DaVe McComb's screenshot shows his AI 'Archer' reporting a loyalty-chat win: Verizon offered 25% off plan charges — about $33.75/month for 12 months, no plan change or contract — which, with a hotspot already cut, drops his bill from $194.74 to about $161/month, roughly $405 a year. Archer asked whether to accept; he replied 'Yes, take the discount', and Archer confirmed it had told Verizon to apply it and would verify.
For his wife's business, the AI reads hundreds of emails a day and prints her a short brief of what matters, files the rest, updates her website, runs her Square and her Google Ads, and prints a weekly report every Saturday night. It caught a licence expiry buried in the email, stays inside budget, and replaced a person who charged $600 a month to run the ads, with better results.
Marty (MartydudeVR, 6.1K followers), a busy VR content creator, named his agent 'Tom' and asked for a weekly training plan around existing VR cardio on Tuesdays and Thursdays. Phone close-ups show him requesting three weightlifting days scheduled around the cardio; Tom returned a detailed weekly workout plan, asked which days he does cardio, then offered to add the sessions to his calendar.
The AI enabled a camera on a test device and he taught it to reboot into the previous kernel on build failure so it could iterate for days; ended with a usable setup and camera login working.
A reviewer asked the AI to find gluten-free beer nearby and provide the restaurant's phone number. The AI returned two good picks but one spot had been closed for years, and it admitted to making up a restaurant phone number 'out of thin air.' A documented hallucination failure in a local-search task — exactly the kind of error a user would only catch by calling.
In a Vox feature on personal agents, entrepreneur Matt Van Horn said his wife sent Instinct a photo of a parking ticket with no instructions at all. The agent, which already had her credit card on file, simply paid it.
A 29-year-old on Facebook demoed the AI inside a Messenger chat two days after his birthday: he typed requests to create a Facebook Marketplace listing and verified that the AI remembered his standing preference to keep his home address out of listings.
He works with collision data as a UK highways engineer but could never answer the simple version for his own commute: which route has actually hurt people. He built SafeRoute with AI, which plans a route and scores it on five years of collision records; colleagues then asked for the same data as branded evidence reports for committees, which the tool now produces in about 30 seconds instead of a day of manual work.
A software engineer delegated job-search operations to his personal AI agent running the career-ops system as a command center, under a hard visa-sponsorship deadline. Over eight weeks it evaluated roles, tailored CVs, tracked 21 serious applications and prepared interviews, ending in 7 interview processes, 2 offers and an accepted role in Amsterdam.
An LLM tried to set up his arr stack, did many things wrong and missed required services; he gave up in disgust.
A Dutch man brought his own AI assistant, Eva, running on the AI, into a live call with his provider Ziggo. Eva had compared a rival offer against his contract, and spoke during the call when he asked her a question. Inside 30 minutes he had a renewed 24 month contract that cuts his total bill by 1,250 euros.
He built an app with AI that reads his solar inverter and batteries, forecasts tomorrow's solar generation and household use, and emails him a report. It sets the battery charge level for the overnight cheap tariff based on the forecast.
He wanted one place where people can find all the apps he is working on. The AI built the site, beardfm.app, in about 15 minutes.
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.