Andy Bennett built Helm, a Mac, iPhone and iPad job tracker with CV tailoring, with AI writing all of the code and most of the design across 641 commits and 21 plans. The first working Mac version landed in about eight hours of commits on September 11; both apps are now in TestFlight beta.
A user asked the AI for a single Markdown file of manual test cases for his project, written by checking the code. Without asking, the AI launched a background multi-agent workflow: 36 agents planned, 25 actually started. It ran until his usage was gone, the agents failed, and the session ended. No verifier finished and no file was ever written to his project. The AI later admitted one pass would have done the job.
This one went wrong. See what happened before you try it.
Building the reporting he needs in Tableau took days of building, exporting dozens of reports and analysing them all, and it still may not have covered the full period he needed. With an AI connector, the same job was done in 10 minutes.
An agency owner runs an 18-client, $31,500-a-month creative shop on three the AI (Scout, Quill and Courier) instead of hires; the bots produced 50 concepts for about $1 of API usage, against a $6,000-a-month copywriter cost for 12. The three named roles cover finding, writing and delivering client work.
The same business owner runs AI on a weekly cycle for search traffic. It scans competitor sites for the keywords he tracks, compares his own pages against them, suggests changes, and finds and puts in internal linking fixes itself.
His agent researched a prospect and sent the email. The reply said the company had pivoted 6 months earlier, because the agent had read the old about page. The research saved him time, but the stale facts cost him the reply.
This one went wrong. See what happened before you try it.
A Japanese Threads creator split his account's operation into six AI agents with one deliverable each (researcher, writer, poster, fetcher, analyst, supervisor) and reports 1,181+ posts and over 10 million cumulative impressions. The writer agent may only cite numbers stored in a single 'truth file' so it cannot inflate results, the poster must confirm the published URL rather than the button click, and the fetcher records 'not retrieved' instead of zero when a fetch fails.
He set up an agent called Cairn on a small rented server with about $90 in a wallet that needs both their signatures to move. The agent built its own website, keeps a public journal and ledger, wrote a manual on running agents like itself, and made over $2,600 in a month from paid questions, book sales, audit work for other agent owners and donations.
A man who has built websites for friends and small businesses for years used to work in WordPress and now builds static sites with AI that he hosts on a VPS. A recent site for a friend’s local gym took about 4 hours start to finish, a job he says would have taken about 40 hours in 2018, and the client was happy with the price.