An agent migrated his whole setup to a declarative system, backups included
Migrated a fairly complex setup to NixOS 1 to 1 with an LLM agent and gained declarative services, backups and redeploy tooling.
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.
Migrated a fairly complex setup to NixOS 1 to 1 with an LLM agent and gained declarative services, backups and redeploy tooling.
A commenter says he feeds the AI his booking details and complaint and lets it negotiate petty corporate disputes. For a Milan hotel stay, the AI first won a free night he rejected, then a 25% reimbursement he accepted.
He got a Psion Series 5 and a USB to RS232 adapter working and saw the OPL programs he wrote as an 11 year old run again. He had the AI port some of the OPL code that had changed between the Series 3a and the Series 5.
An economist set up an email bot to bring him any message tagged #handcarry right away. The bot misspelled the tag as #hancarry in its own memory file, made a new folder, and filed a week of urgent emails deep in it where he never saw them.
Another user describes the AI’s phone work. It called AT&T and argued for a discount because his service is poor where he lives, staying on the phone with real people across multiple transfers and departments for 98 minutes, and got the discount. The amount is not stated. His other two cases from the same comment are recorded separately where they stand on their own.
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.
A teacher built his own LMS layer on top of Google Classroom using AI and Apps Script (he names the AI and the AI as both able to drive it): far deeper grading features than Classroom offers, a quick-select comment bank where any feedback he writes can be saved for reuse with one click, and on return each student automatically gets a formatted read-only feedback document both emailed to them and attached to the Classroom assignment. Student data never leaves the district's Google ecosystem.
A 10-slide head-to-head test gave the AI and the AI the same job: a table for 4 in downtown Denver, Friday Oct 9 at 7pm, under $40 per person, good vegetarian options — verify availability without booking and draft an invite. The AI checked only OpenTable, declared no bookings and switched restaurants; the AI confirmed via City O' City's own Toast page that a 7pm table was bookable, which the tester verified manually (The AI took ~6 minutes vs the AI’s ~3).
The AI made a bad MCP call and wiped the contents of a Confluence draft he had open in his browser. He lost two hours of work.
Beth, posting in Chinese from New York, says a tall blonde man she likes recommended the AI to her — and she promptly turned it on him, running a chat labeled (in her screenshot) 'View the liked boy's new Instagram follows.' the AI’s analysis showed that although he claimed to have cut ties with his ex as of April 2026, he is still following her and never unfollowed, and had minimized four to five months of reconnection as 'contacted a bit a few months ago.' She notes this kind of follow/unfollow tr
A man put the local tax law, the tax guide, and his and his wife’s documents, bank statements and income statements into one folder and asked the AI to fill out their 2025 Swiss taxes. He says there was nothing to fix, and he submitted it.
The AI vibe-coded a native mxswitch app running on both laptops that sends a DDC command to switch monitor inputs when keyboard and mouse switch devices.
Since the AI got access to his Strava through an MCP, it has become his running coach. He says his progress is clear and he may reach 150 km of running in one week, but the sessions are so strict he feels like he is doing maths.
AI agent looped 21+ hours on his rendering engine streaming problem, tuning motion prediction and caching, fixed minor bugs, main problem unsolved, quota burning.
The AI takes his scanned documents and books them straight into Money, his accounting software.
The developer building the signup platform for his wife's education seminars stopped answering emails, and they had no access to the code. With no money to hire another developer, he rebuilt the platform from scratch with AI as his coding partner, and it is running now.
Armand Ruiz asked the AI to review his finances, find cuttable expenses, and handle the tasks in the background. The AI recovered $2,120.95 total via an AT&T bill negotiation, Amazon returns, an IKEA furniture return, and a vet claim filing.
A services co-op connected the AI to its Salesforce. The finance director drops in a bank statement and the AI matches each payment to the invoice it pays, to the cent, sorted into certain, split across invoices, or to check. Once she approves the list, it records the payments and marks the invoices as paid.
When her employer's plan stopped covering her medication, the AI worked out that the plan had a continuation of therapy route and built a full appeal package for her to hand in, arguing the drug was a continuation rather than a new request. She filed it, and her doctor used the letter template it wrote.
Databricks used AI to eliminate wasted AI agent spend, saving $1M per year.
His corals died but the tank lights still ran a coral schedule. The AI drove a Raspberry Pi over Bluetooth, changed the light and pump schedules, and found the lights had been dimmed to about 35 percent and their clock was 6 and a half hours late, then fixed the time.
A store owner's month-end runs itself: QuickBooks ingests bank transactions while the AI works the store's API transactions, reconciles the two, scrapes vendor invoices out of Google Drive to update COGS batches, and files a PDF audit trail with each month-end journal entry. Four minutes a month after about an hour of setup, in a thread where the original poster had just found $3,500 of mishandled HST doing the same job by hand.
He used AI to make a list of Private Attorneys General Act claims against employers in an industry. The list was made in support of an organizing drive, and he says the tool is almost always helpful on tasks that involve a lot of data.
Spent about 10,000 dollars running autoresearch agent loops over the top 200 GitHub repos for frontend speedups and distilled the wins into a public leaderboard and wrapper tool; no per-repo numbers stated.
He says the AI automated the heating control in his home. The post tags his heat pump and his Home Assistant setup.
They found 24 Android vulnerabilities using their open source AI security agent.
Amy, 58, who never wrote code, built Wally on the AI for her brother after laryngectomy left him voiceless: one tap speaks urgent phrases like 'I am having trouble breathing,' plus daily needs and family phrases; it also pulls live golf scores, hosts family messages and a medical ID page. He used it daily until his death in August 2026.
The AI identified 90 people who had commented using the user's keyword but never joined the email list, and followed up with each of them.
He connected his Gmail to an AI task that watched every new email and was only allowed to interrupt him when a message needed action or came from family, with a draft reply ready. All afternoon it alerted him three times, for a bank message, a family email he answered from its draft, and a ticket itinerary change he needed to act on.
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.