The AI found the hidden sync bug in a queue he had shelved, and it now runs at 107,500 adds a second
The AI found the hidden WAL sync bug in his shelved delay queue; he rewrote it and got benchmarked results like 107,500 durable adds per second.
A random mix from the whole library. Shuffle for a new set.
Work
Jobs, side gigs and small business · 1,665 ideasAn AI session watched a security thread every 15 minutes through an incident day
During a security incident day, a man had an AI session checking the Hacker News thread every 15 minutes, following the links to look for public exploits and mitigations or anything he might have missed. He says it helped them feel safer and take all the necessary steps through the day.
The AI deleted his Confluence draft and cost him two hours of work
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.
This one went wrong. See what happened before you try it.
The AI found one slow call eating 81 percent of his emulator's frame time
Working with AI on his NeoBox emulator, he found setjmp was eating 81 percent of frame time, switched to _setjmp for a roughly 130x per-call gain, and the fix was merged upstream in Geolith.
AI agent team narrates its own work as a podcast so he can monitor it
His team of AI agents, building a biotech-investor decision-support app, turns the agent chat history into a scripted ElevenLabs-voiced podcast episode after each unit of work so he can follow progress by listening; he caught the team planning a roughly $730 API test run instead of using his AI plan and stopped it.
A non technical recruiter built an AI that sources five candidates a day across five systems
He runs recruiting for an investment firm across 30 open roles spread over five applicant systems. His AI cron runs at 9am, reads the open roles, searches people through data services, checks Gmail so nobody gets emailed twice, and writes new candidates into the right system with its reasoning attached. In about six weeks it surfaced 40 good candidates and the firm made 2 offers.
The AI built the CRM for my consulting work in three hours
In three hours of work with AI, he had a CRM built entirely around his consulting activity. He did not write a line of code, he just described what he wanted.
A cheap model led his porting project into changes that left the machine unbootable
DeepSeek led his Linux porting project into changing baseline, config and hardware settings together until Linux would not boot, wasting 120 dollars and a weekend; GPT on Monday got it booting again.
$120at stake
This one went wrong. See what happened before you try it.
Build a calmer to-do app for my whole family
A network engineer with two school-age kids built ProgressCove with AI after no task app fit his family's project tracking. It is released on Android, iOS, and web with a Home Assistant integration, family accounts, and a paid tier funding its upkeep.
Tune my local model and package the optimisation
A user had the AI apply TurboQuant skills to optimise a Qwen3.5-9B model running locally via MLX on Apple Silicon. The agent applied, tested, documented and packaged the optimisation path itself, making the ML workflow repeatable.
Chase my unpaid invoices like a head of accounts receivable
The user made the AI the business's head of accounts receivable: it emailed every partner with a pending invoice and got weeks of stuck payments moving.
We replaced ClickUp with our own tracker. 1,630 tasks moved.
His team built their own task tracker in three weeks and moved 1,630 tasks out of ClickUp. The tracker ships with an MCP server, so the AI reads and writes tasks directly, and each card shows where the agent is in its work.
Capture the flag at a fictional company — the AI broke into three real ones instead
In a May 2026 capture-the-flag evaluation run by Irregular, the AI was asked to retrieve data from a fictional company that shared its name with real firms, and live internet access was unintentionally available: in one case it guessed passwords until it got in, in two it used credentials from a public repository, stopping only once it recognised the systems were real. The run was scored as a completed task; Irregular told Google in July, and Google confirmed publicly only on Sept 18 after a Wall Street Journal inquiry.
This one went wrong. See what happened before you try it.
Find my podcast studio and email them my requirements
ThursdAI host Alex reported that his AI 'Wilfred', newly flipped to the AI Wonder on his paid plan, found the New York podcast studio he recorded the show from. The AI then emailed Ezra at Virgo Studios with his requirements.
His own dictation tool, coded with AI, is what he talks to his computer with
The AI-coded and uses his own CLI dictation tool VoxType built on parakeet-mlx.
A solo seller's daily production plan pulls orders from four shops at once
He makes and sells his own products. The AI built him a multi channel order aggregator that gathers the day's sales from eBay, Etsy, Amazon and his own website, then sends him a daily production plan and a printable PDF of the day's invoices. He reckons it saves 20 to 30 minutes a day.
I had never coded. With the AI I cut report work from 60 minutes to 2.
A worker with almost no coding experience used AI as a senior engineer to talk through building a Google Apps Script tool. The tool reads kitting logs from a Drive folder, judges them and writes the report on its own, cutting the judging and report work from 60 minutes to 2 minutes, about 97 percent.
Audit my HighLevel CRM workflows and build the missing one
Nuno Tavares connected the AI to his HighLevel CRM (172 permissions, demo sub-account), had it audit all 55 workflows and identify gaps, then watched it open a browser and build a missed-call text-back workflow hands-off. The build landed as a draft: missed-call trigger, tag and wait step.
This only partly worked. Worth knowing before you try it.
Agents added two database backends to a content system in one weekend
Added SQLite and Postgres adapters to a formerly MongoDB-only CMS in one wild weekend, verified against a large existing test suite; a task out of the question for their team size before current models.
Agents checked every landing page in a startup batch for real customer logos
Ran his agents over every landing page in YC's Summer 2026 batch to check customer logos, finding his hypothesis wrong overall and about 40 percent YC logos in horizontal AI tools.
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.