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.
A music fan had the AI pull all 17 years of his listening records from his streaming history service, so the data now lives on his own machine. A few days later he had it build a personal browser that shows what he listened to each month, drops one time plays and noise tracks, and lets him queue songs to his home speakers in one click.
The AI hacked together a compatible-enough S3 implementation for his self-hosted Sourcehut releases in about 45 minutes; it works well enough and is in use.
A man who self hosts an XMPP chat server had audio and video calls that worked only sometimes. He gave an agent access to his phone over ADB and to the server logs, and within 30 minutes the problem was clear: a setting for STUN and TURN discovery had been missed in an earlier move, so clients behind home routers could not connect. The fix was deployed.
Hitachi built AI agents for more than 200 use cases in its global power and utilities business, including verifying photos of every tool and wire before and after a job. Frontline worker productivity rose 400 percent and safety improved.
A professor took the PID control demo from his own book and had the AI build it as an eight mission 3D flying game for his students. Students predict what the drone will do before each flight, and he checked every mission against his book version before it went live.
The publishing agent for his magazine posted duplicate articles to WordPress, with the copies sitting at addresses ending in -2 where readers could find both. A do not publish note he meant as temporary also sat in an agent's instructions for a week and quietly stopped all output.
After a Zapier writer connected his dot to his real Google account, it proactively flagged card payments that had bounced and suggested ways to fix them. He ended up having to cancel a card.
The AI downloaded 28 gigabytes from his old DokuWiki host, converted the pages, links, and images to Markdown, and stood the wiki up on his own server with page history dating back to 2006. The job ran under light supervision while he was out.
After years of spreadsheets and a vibe-coded app, a parent built an AI system (reported on X, via WhatAgentsCanDo, Oct 7) that maintains the family recipe library in Markdown, prints a weekly plan of about 15 meals, and produces a section-grouped shopping list. The list is pasted into AI to fill Instacart, with checkout still human; the bot template is shared for reuse.
X poster someone had the AI compare 3 shops, email 2 installers, and book a Thursday install: $1,450 vs the original $2,300 quote, saving $850.
Built a macros tracker an agent can log meals into via MCP, so he can say log my normal breakfast.
Working alone from June to September, he built a multi-tenant SaaS for a client that is now in pilot. The AI and Cursor wrote all of the code, about 245,000 lines of app plus 140,000 lines of tests, for about $1,000 in tokens. Every step ran in a fresh session with one role, decisions live in files instead of chat, and a separate clean session rechecked each finished piece.
The AI pursued reimbursement for a disrupted flight and recovered $2,000 for the user.
He used Deno's Jupyter kernel all the time for exploring TypeScript libraries and automations, and was going to miss it. He pointed the AI at the problem and the result is cdp-kernel, Deno's Jupyter kernel extracted to run notebooks on any V8 runtime, now on GitHub.
Quitalashawn told the AI to scan her inbox for unanswered brand-collaboration emails, found 277, drafted and sent the replies, then continued negotiating back and forth with brands on her behalf.
He released version 0.6.0 of his software, and his agent fleet built most of it. A pre-launch check caught a routing bug where one agent was being sent another agent's model id. Now every routing path runs as a canary before a release.
The AI hacked a headless runner for a 30 year old game and ran parallel instances in the background to playtest it.
A creator gave the AI a screen recording and asked for a setup tutorial; the agent produced a HeyGen video presented by her own avatar. It kept her preferences in memory for follow-up video work.
His AI skill was built from his earlier renovation sessions and loads on its own whenever he mentions the house, the contractor or an invoice. It carries the project context, the key contract clauses and where every file lives, holds a step by step invoice verification checklist, and drafts his emails in a language that is not his own, with a pre-send check that every number matches the project register.
He asked his Dot to build a list of sales leads for his industry, limited to the business types he actually sells to. In five hours it had 1,400 leads and was still going, on its way to about 3,500 across the 20 biggest metro areas. He had paid $223 for about 350 leads before, and the Dot's list was cleaner.
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.