An LLM used his code editor's language server to fix a few dozen bugs in a Python rewrite
LLM used his LSP on a Python-to-ucode rewrite and fixed a few dozen bugs from interpreter/LSP errors (work deliverable).
348 new ideas this week
Level up how you use AI
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.
My agent builds my monthly finance statement from my accounts in 15 minutes
For about 25 years he finished a monthly financial statement in Excel by hand, and it took hours. He connected the AI read only to all his accounts through Plaid. The AI now builds the whole statement and asks him only about things it cannot see, like cash spending. The job takes 15 minutes instead of hours. He is having the AI build him a custom money dashboard next.
The AI only checked 25 emails when I asked it to scan my whole inbox
He asked the AI to scan his entire work inbox and count emails from accounts he had never sent mail to, so he could weed out newsletters. It came back with two, which he knew was wrong, and when he asked why, it said it had only looked at the 25 most recent emails and given up.
This one went wrong. See what happened before you try it.
My scheduled post went out on its own to three networks
He set a post up in an AI app and his small tool crier sent it at the set time to Mastodon, Bluesky and Nectvia. Nothing was running on his side when it went out.
Manager agents review every sales demo call and coach the humans who ran them
AI manager agents review transcripts of every demo call, compile insights, update company docs and give feedback to the humans who ran the demos; agents own KPIs and escalate to humans.
Fight every company that shorts me — refunds, appeals, warranties
Over seven months a parent ran every consumer dispute through AI: it researches the contract, marketing promises, regulator and owner reports into a dossier with a three-branch plan, files complaint forms through the Chrome extension with photos attached, and coaches him live on support calls from transcripts he feeds back. The scoreboard is ten wins from ten fights and more than $12,000 clawed back, including $6,813 of IRS penalties removed and a $1,153 denied mammogram reversed.
$12,000
We dropped Power BI. The finance team builds its own reports with AI now.
He says his business got serious about Power BI reporting about twelve months ago, then three months ago threw it out and replaced it with AI licences. A third party warehouse tidies the data with an MCP on top, and the finance team now builds its own reports quickly instead of waiting on a reporting team.
His agent emailed and phoned Columbia about a return. The call died when it was transferred.
He wanted to return pants to Columbia. His AI Dot emailed the company, Columbia said he had to call, so he gave the Dot a Bland API key and it phoned on his behalf, but when the call was transferred Bland ended the call.
This only partly worked. Worth knowing before you try it.
Watch Autotrader and alert me to mispriced supercars
MindStudio tested an AI monitor that watches Autotrader listings and flags cars priced well under market. Scheduled checks deliver deal alerts to Telegram, and a continuity setting stops it re-alerting on the same listing.
An LLM built the sandbox a big project had left undone for almost a decade
An LLM built Homebrew's Landlock Linux sandbox that had sat unfixed on the roadmap for almost a decade, reviewed line by line by maintainers and shipped.
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.