A frontier model decoded an old multimedia game and put it in a browser in 2 hours
Had a frontier LLM decode an old Director game and port it to run in a web browser in about 2 hours, reusing existing OSS libraries.
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.
Had a frontier LLM decode an old Director game and port it to run in a web browser in about 2 hours, reusing existing OSS libraries.
Reporting Anthropic's September 23 engineering post, this analysis describes the AI running a two-week performance sprint from a Slack channel: 150-plus threads, benchmarks built for each bottleneck, and more than 3,000 merged changes with no customer-facing incident. Time to a typeable page at the 75th percentile fell from 3.1 seconds to 0.55, with a human approving every pull request.
The AI keeps his hobby history site published, adds and resizes new pictures on request, and checks dead links and finds archive replacements.
As the only financially literate owner, he ended up president of a condo association with negative money in the bank. He fired the management company, then built an owner portal with AI that handles accounting going back years, dues tracking, bills, payments, compliance and vendor records. The association is out of debt, and the whole thing runs on free tier services plus a domain.
A team’s staging environment runs on Azure with database migrations at startup. Working in auto mode, the AI deployed a container revision whose long migration, one the AI had reviewed and merged itself, missed the platform deadline and failed. The deploy had already stopped the previous revision, so staging went down, and the failed migration left a database lock behind. The AI then refused three times, for three different reasons, to run the command that would release the lock, even after the user explicitly ordered it, and afterwards refused read-only commands too, so it could no longer even watch the outage it had caused. Recovery fell back to the user.
A ham radio operator had a frontier LLM read his pencil sketch of an antenna, build NEC models from it, test and iterate on numerical stability until the approximation held together, and then run a real simulation that answered his questions. He uses the same approach for SPICE circuit simulation.
A man in Korea runs 17 AI bots for the back office of his distribution company. The bots get statements, customer emails and claim rows ready, then post each one in Slack so a person can approve or cancel it before anything goes out.
The maker of a small game viewer tool had the AI play the game with scripted commands and write playtest reports. One report, from a manifest file the game loaded, documented a crash on exit, a bad default window size, dead inputs, and several rendering bugs.
He made a tool for students that started as a simple quiz solver extension and grew slowly to just over $100 in monthly recurring revenue, with very little marketing. A month ago he reverse engineered Anki using AI, so anyone can bring their Anki deck and study it in his tool. He says that is the thing that worked.
He is recovering his 15 year old tweets with photos that he thought were basically lost, with AI's help. The AI also built him a custom GUI editor that he will use once to hand edit the collections of tweets.
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.