A teacher and dad rebuilt his classroom card-review game into 'Bad Mascots', a mascot-horror study game for his 2nd-grade son: The AI planned the conversion, wrote the image prompts and built and tested the game file, questions come from a photo of the boy's weekly vocabulary sheet, and progress saves to its own Google Sheet. Weekly updates take about 10 minutes from photo to approved questions, and the game is live and playable online.
In the same CEO's US business line, an agent reviews the call-center operators' calls, transcripts and CRM results every day, then writes the human supervisor's task list with numbers and deadlines and sends each operator 3–4 small skills to fix from their own recent calls. Each skill is marked pass or fail every evening and closes after three consecutive passes; the agent only suggests — people decisions stay with people.
The third case in the same comment: he had the AI call three local companies to get quotes for a fence at his property. Quote gathering for home work normally means three rounds of calls, messages and waiting; his agent made the calls itself. Prices are not stated in the comment.
His racing game across Tower Bridge and through London is live on his own server, playable in the browser with no install. He built it with AI, including an AI opponent he says he cannot beat, and the live game has 7 AI cars, 3 laps and a hard difficulty setting.
A moderator shared another person's result: their AI searched for money owed to them and found over $1,500 in unclaimed funds across two states they used to live in. The prompt they shared also has the AI check the claim status every 5 days and report any change.
He has the AI read and search across his email addresses and scan a few times a day for mail from people who matter, like customers and family. He has basically stopped checking email forty times a day. It never sends anything; sometimes it drafts replies, which he rewrites in his own words.
A musician in a home studio asked the AI to search his Instagram followers and follows for musicians active in the past year, select two songs per musician, and compile them into a YouTube playlist. The agent succeeded, creating a 6+ hour 'Community Listening' playlist — music mostly new to him from his own network.
A directory-listed the AI called Haus Hunter runs its owner’s ongoing home search with real OR and IF/THEN rules, checks listing photos for unwritten details, and weighs commute and local context for each candidate. It investigates homes the way a good buyer’s agent would, continuously.
A blind runner has the AI list and add his Achilles running-group shifts in Rosterfy, a platform he finds largely unusable on iOS, simply by sending a WhatsApp message asking to be signed up.
Filing travel reimbursement, the author realized he never got his hotel receipt. His AI 'Tofu' emailed the hotel, then called them two days later — and the hotel sent the receipt while he was still on the line.
Ekaterina Li (someone) ran AI as a full personal assistant for one week, delegating eight tasks from idea to execution. In the thread's comments she adds that she pays $20/month for the AI and considers it sufficient for everything she asked.
The AI CEO agent and sub-agents run one of his side hustles daily, pitch ideas each morning, and implement the chosen one via PRs; revenue up 4x in the past month.
A creator building a pink DIY bag charm that shows time and weather had the AI source, compare and — after her approval — order an OLED screen, ESP32 board and LiPo battery. The finished charm is shown working on her handbag.
In his daily bioinformatics work, a volcano plot the AI’s script made had a subtitle saying 57 down while the blue side of the plot showed thousands of points. He told the AI to find out why, and the AI traced it to an inner join that pulled the up and down counts from a different comparison than the one plotted, then admitted the mistake.
This one went wrong. See what happened before you try it.
Her mom normally needs a care service to take off her compression stockings. With the AI she found a way her mom can do it alone in an emergency, her mom tried it and it worked, and her brother no longer has to come over when the care service fails.
The AI wrote a POSIX vi implementation from the POSIX specs over a weekend; he has used it for over a month with no fault found in its POSIX implementation.
Three independent AI audits went through his Cashu payment server and wallet library line by line for a new release. All of them found real bugs that previous audits and models had missed, and one model found a higher severity bug nobody had caught. The release fixes what they found.
The AI went through his setup and found years of pending uploads that were never being cleared, which saved $2,000 a month. It also found more than $1,000 a month in S3 transfer costs to imgproxy, and moving that work from Railway to Fargate makes the transfer free.
Asked to move a client meeting to the following week and follow up on a spreadsheet, Toby assumed the owner used Gmail when his setup was Zoho with Apple Mail and Outlook; a setup widget failed and it sent three malformed links. Neither the email nor the rescheduling was completed, though it did warn the spreadsheet looked view-only for the client.
This one went wrong. See what happened before you try it.
He built a local video studio setup where the AI does the actual editing from the terminal on his Mac: it cuts, plans the edit, writes captions and checks the result. He gives feedback and approves, and the setup uses Remotion, ffmpeg and local Hebrew speech to text.
A shopper who never used to bother chasing refunds now hands the AI every bad experience - crappy Instacart produce, a heat-press controller reading the wrong temperature - and the AI opens the case, tracks it, and responds to support until it is resolved.
She wanted to clean up her brand assets folder but did not know which images her site still used. The AI matched 242 files to the site by the fingerprint each file carries, then deleted or archived the old ones that were no longer needed.
His open source agent wakes on its own every few hours and studies new tokens on the Base blockchain, checking each one at launch and again at 1, 6 and 24 hours. Of 1,045 new tokens with at least $1,000 of liquidity, only 12 still had it after 24 hours, and all 897 that launched on one exchange type were drained.
A user running the AI in manual mode, where every edit needs approval, found the agent rewriting rule files and creating a script without confirming anything. The model had read the session's permission mode once at the start, when it still showed 'default', had no way to see it change, and when challenged answered from that stale reading and guesswork instead of checking.
This only partly worked. Worth knowing before you try it.
He gave the AI his 2022 blog post about building a neural network and asked for a manim animation video with narration. The AI made the video in about 55 minutes, and it also spotted bugs in the original post while converting it.
His six publishing agents used to wake on timers about 20 times a day, and most wake ups had no work to do. Switching to waking only when a task arrives cut the daily bill from $5.86 to $1.21, about 79% less, and September's four long form articles cost $50.53 total.
Each week he drops three files into a folder: an Apple Health export, his Peloton data, and a spreadsheet of extras. The AI runs the numbers through 19 metrics, flags shifts like cycling power dipping after a hard hike, and updates a dashboard he checks to adjust his training.
He built an outreach setup last month. The AI finds leads, checks them, writes emails, sends them through AgentMail and tracks replies, and he reviews the work once a day.
He forgot to charge the batteries for his road bike's electronic shifting before a ride last week. After a weekend of back and forth with AI, he now has a dashboard that can alert him when things need to be charged.