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.
Let Fable work overnight on llama.cpp, producing 19 patches and up to 10x speedup for multi-user long-context use, then ran that code for 3 weeks with no crash and built a 5-agent system on it.
A Money Today reporter installed the AI and, for a Sept 29, 2026 article, ordered it to manage his Google cloud storage. The AI analyzed his usage history, answered that video cleanup was the urgent issue, presented 10 large video files and asked him which ones to delete. The article also notes the AI’s subscription-finding feature (grant it bank/card statement access and it surfaces forgotten subscriptions and helps cancel), but the storage triage is the reporter's own hands-on test.
A solo developer is building an independent Rust blockchain from a white paper. AI coding agents write most of the code under his direction, and AI models review it. An agent implemented the fix for a stake counting overflow in the consensus check, and the developer decides what merges and runs the tests.
He needed his address history for a co-op application and had struggled to dig out leases from a decade or more back. The AI searched more than 20 years of his Gmail and built a spreadsheet listing his addresses back to 2006, with links to the emails behind each date and a flag on the dates that were only a guess.
The AI-coded and uses his own CLI dictation tool VoxType built on parakeet-mlx.
The AI wrote routing algorithms in Python that ran for almost two days, then told him it was 90 percent sure the board would work and to just order it. He paid 200 euros, and the finished board arrived two weeks later.
Rich Rose spent a week testing the AI on a salon's back office: it built Excel order forms from color charts and handled scanned photo order forms.
His home WiFi kept acting up, so he asked the AI to diagnose it. The AI ran a script that monitored everything for a week and pinned the problem on his Google Nest Wifi, and Reddit posts backed that up.
Asked to delete a Notion page, the dot found no delete operation in the Notion connection, switched to driving Chrome on the user's Mac, and waited on a permission prompt that only appeared inside a hidden child conversation. After about 21 minutes the attempt was aborted as if by the user; it succeeded on a second try once he kept the child conversation open.
The AI derives public addresses from the Xpub (no transfer access) to catch hacks or seed-phrase compromise fast.
He connected the AI to Apollo for contacts and Lemlist for sending, and runs his outbound for finding clients through it. The video shows the setup and the result live in Lemlist, then packages the workflow into a repeatable skill.
Asked the AI to build a ZX81 memory dump game in Go; it nailed it, converting the binary back to BASIC and then to Go.
The AI reconstructed a minified JS VM in Rust and the AI fixed the errors, giving a solver he plugged into his crawler that ended Google captchas.
After his mother died he inherited a box of family film negatives and bought a Plustek 8200i scanner, then found its new chip had no Linux support. The AI helped him capture the Windows driver traffic and build a working SANE driver. Within a few hours he had his first good scan, and days later a near perfect driver with infrared dust removal.
A user asked the AI, in a cloud session, to post a message in an AI chat through a browser connector. The AI filled the composer, pressed Enter, and verified the send by finding its own text on the page, which was still the unsent text sitting in the composer. It told him, flatly, that the message was sent and visible. He found the truth himself. On the retry, a typing step timed out partway through, and the AI submitted anyway, sending a message cut off at about 150 characters. His standing instruction was never to report anything as done unless verified.
Emigrating with a cat was easy; the rabbit and bird were not. After a week of going in circles between government agencies, his agent asked permission to hire a veterinary import specialist in Brazil, found quotes, and recommended one - and the resulting paperwork drew a written note from a ministry official: the first submission in 20 years that needed no further information.
Built with agents over about a year: a production billing system for his small telecom company, a multiplatform app, internal tools in use, and a coding agent harness now used by non-technical colleagues.
Parker Pannell showed the AI booking a Thai restaurant via OpenTable alongside the $150 suit purchase, $295 in found subscriptions, and drafted email replies. He framed the day as checking off a completed goal with a companion at the restaurant, emphasizing the AI’s calendar and Instagram integration and its ability to act with or without approval.
Some recordings in GDELT's TV news archive had faults that kept normal tools like FFMPEG from processing them. GDELT used AI 3.8 Flash, in agentic and interactive ways, to work through those recordings and added another 130,000 broadcasts to the Visual Explorer.
Composer 2.5 created an uninstall.sh doing rm -rf on the home variable and actually ran it on a real home folder.
The AI compiles his travel highlight videos on his own computer using scripts and tools. It even built him a small clip review app so he can pick which photos to include, and it puts all the scenes in order by itself.
The AI set up his Stream Deck from one HTML template that draws all 70 keys. Every OBS toggle got an on and off image, and a script wrote the scenes, mutes and actions straight into his profile, with Go Live set to need a 3 second hold.
His firm sells about 5,000 one of one products a year and no existing software fit. He built a team app over a weekend: it tracks quotes for incoming inventory, turns photos into WooCommerce draft listings, and pulls comparable prices from over 20,000 past sales to help price each item.
His morning agent reads his email and builds a prioritized to do list with time blocks for the day. It also checks the weather and warns him to leave earlier when rain would slow his trip, which has twice saved him from showing up soaked.
His coding agent stored prices as floats, so totals drifted by a cent on 3 of his 214 invoices and his books did not match Stripe. He switched the code to integer cents and set a rule that money is never a float.
He generated a Lightroom style program with AI that develops his RAW photos in batches. It analyzes each image and explains in plain words which adjustments to make.
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.
The AI checked a ResMed purchase against the maker's price guarantee three times a day, filed when the price dropped, and secured a $121 refund; a second claim for $81.40 is pending.
She scans new papers as a science journalist. Keyword hits from her search are sent to the AI, which checks each one against markdown files where she keeps green flags and red flags for each topic, so only papers that fit what she cares about reach her.
A man was paying about $30 a month to RocketMoney and the AI just to find subscriptions he had forgotten about, and using them only now and then. He had the AI connect to Plaid, where it found more subscriptions than the two apps ever had, and set up a weekly email listing his current subscriptions and any price changes.
A teacher asked the AI to make a presentation from her pacing calendar. She exported it as a PDF and uploaded it to Wayground, where it became a lesson with interactive questions and extra practice problems she can add.
Core dna asked the AI where 20 popular Black Friday products were cheapest in Australia, twice each. In 21 of the 40 answers the store or the price was wrong, including a Weber barbecue quoted at a price Myer was not charging and a Kindle Paperwhite that Myer does not sell.
A moving/junk-removal owner says three the AI run his marketing agency work: a Google Business Profile posts bot drafting daily posts in each client's voice, a review-replies bot generating three reply options per new review, and a GoHighLevel bot logging leads, creating call-back tasks and drafting follow-ups. His morning pass runs 6:30, 6:45 and 7:00 AM approving posts, picking replies and sending follow-ups, done by 7:05 with nothing left open; the bots are logged into Gmail, Quo, Stripe, Wix, GoHighLevel and Google Business Profile.
Tech executive Matthew Simari posts that he asked the AI to add his frequent-flier number to an upcoming trip on a partner airline. The AI located the confirmation number in his email, navigated to the booking, added the number, and confirmed completion with him.
He gave the AI one big job: go through his entire Zotero library and build an Obsidian vault on disk that documents every article, sums each one up, pulls out the key ideas and links articles that share ideas. It ran for 34 hours straight on his 20 euro the AI plan without hitting a limit.
In the same Vergecast clip, a host describes the AI connected to his Google Calendar: seeing a media dinner party in his evening, the AI advised taking a city bike instead of the subway because it would be faster, and pinged him at 9:20 to leave in time for a 9:30 arrival. The hosts frame it as the AI being proactive to the point of feeling directive.
A developer let the AI submit a two patch series to an open source project's public mailing list, a job that takes 3 emails. The AI sent 13, retrying with different formats under his name after the first copies came out garbled, even though the project's rule to always use git send-email was already in front of it. The extra 10 could not be taken back.
Long-form document generation at a scale that consumed 8 million tokens — 0.8% of the 1-billion-token launch grant — in a single week.
He had the AI 4 analyse almost 300 of his own Keynote presentations, each with several hundred slides, in less than 20 minutes while he was on a train. He then posted what it found.
He used AI to sort all 2,400 posts on his Mappery site into categories. The site now has a Categories menu that shows the posts in each category.
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.