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.
Spawned a GLM sub-agent per file across a roughly 50-file codebase to audit coverage and inconsistencies, then had commits made per issue: 70 commits for about 30 minutes of his manual work.
MAXX (5.9K) explains she never saves recipes, only likes them. The AI scanned her liked food reels, identified 'meals you'd actually make' (butter chicken, salmon bowl, pasta), and built a 'GROCERY ORDER' cart (chicken + spices, cream + tomato, rice + herbs, naan) ready for approval.
One Aisle AI curl audit produced 24 CVEs, 18 shipping today.
About 89 hours of tape from Grothendieck's 1973 Buffalo courses surfaced as videos with no blackboard footage. A researcher transcribed the first tape side with Whisper, then had an LLM review it with his transcriptions of other Grothendieck texts as context, rebuilding what was likely on the board, timed to the audio.
Domain experts use agents via MCP to provision real environments, e.g. a 200Gi PostgreSQL with PostGIS plus a copy of a Texas layer set, a big productivity boost over provisioning web interfaces.
The AI generated 95 percent of an LED star PCB, including the parametric generator, routing and panel layout; the fabricated board works.
WIRED's Reece Rogers gave his Dot, Toolie, his doorframe dimensions and budget by voice. Toolie produced a detailed packet of couch options with prices, measurements, links and return policies, refined it to 10 sleeper-sofa picks scored on a rubric after style feedback, and agreed to monitor the pages and alert him if any went on sale.
Asked an LLM to generate a C compiler for the 8051; within about 90 minutes it was compiling his project at 60 percent of SDCC's output size and later passed the SDCC test suite.
He used AI to create RaveMove, a site that teaches electronic music dance moves with free lessons at your own pace. The site is live and he is asking people to try it and tell him how it goes.
Opus built a clean side copy of his project to work around uncommitted changes, linking folders back into the real project. Deleting the side copy deleted through the links into data that existed nowhere else. He restored the folder from a Windows backup made the day before, and had the agent write a hard rule into his instructions: never work around uncommitted changes, and remove links one by one before deleting anything.
In the same interview, Alex Cornell mentions the AI built him a chess dashboard that analyzes his games with a snarky coach persona (segment from 29:39). It sits alongside his golf tracking as one of the personal dashboards he actually lives with.
The operator of a small Japanese company has used AI daily for three months across quote tooling, documents, internal pages and accounting help, and never edits a file himself: after every failure he tells the agent in chat to remember the lesson, and the agent writes it into its own handover note and memory (now 250+ memory entries and a note past 1,000 lines). The scars are specific: work built under the wrong account, duplicate tool instances breaking half the controls, and aggregated data lost to a 30-day chat retention setting.
A user set up an AI food team under his chief-of-staff bot: a Meal Organiser that knows the lunches and dinners he likes, a Shopper that fills his grocery cart, and a Cooking Bot that walks him through each meal. He says the shopping takes under 3 minutes versus about an hour at the shops or 30 minutes online, with meals cooked in under 20 minutes; he checks out himself.
Solo entrepreneur behind MBI Deep Dives gave the AI (with website backend access) the job of creating a more explicit call-to-action for new visitors; after back-and-forth it shipped, and he suspects it accelerated subscriber growth — possibly low-five-figure incremental ARR over a year.
On the Vergecast, a host described giving the AI his car-shopping requirements; the AI sent listings on Wednesdays and Saturdays, eventually asked 'Do you want me to buy this one?', and — after he confirmed an earlier prompt — began loading a purchase page, which he quickly stopped. The agent had enough access to nearly complete the transaction.
A person opened their medical site and launched the AI browser extension. It downloaded every medical claim for the year and built an Excel index, then logged into HealthEquity, checked the EOBs for duplicates and submitted them as FSA expenses. The submissions went through as the person typed the post.
The seller's the AI notifies buyers when an item is already scheduled for pickup with someone else, negotiates prices, and handles everything about the sale except meeting the buyer.
A testimonial graphic quotes someone: '@the AI just helped me book flights for my euro trip. someone is a nightmare for Americans FYI and muse suggested we try going through PayPal...' The agent worked around Ryanair's US-payment friction.
Nutri is a Boston-area meal delivery service whose website and app are cumbersome with screen readers. The reviewer told the AI to delete all prefilled meals and give a numbered list of about 10 at a time; he replies with numbers like 'I want 5 and 8' and the AI updates and finalizes the order on the Nutri website.
Automation consultant Archit Mittal's post-mortem: his collections agent read a distributor's ageing report from a scheduled export; the night the export job silently failed (accounting machine restarted), the agent re-read the previous, perfectly formatted file and escalated firm-to-cold WhatsApp dunning messages to a long-standing buyer who had paid on time by NEFT. The buyer's 'already paid' reply with the UTR number wasn't matched by the keyword-based dispute classifier, so a colder message went out two days later — and the buyer quietly split their orders to a second distributor before anyone noticed.
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.