A random mix from the whole library. Shuffle for a new set.
Creator Matthew Berman describes giving an AI access to his local machine on a standing weekly job: it scans every file, finds deletion candidates totalling about 380GB, and sorts them into low, medium and high risk. The bot operates under one hard rule — it may categorize but must never delete anything itself.
He spent an evening trying to teach the AI to drive a car. It kept warning that it was about to crash and stopping, even with an obstacle 30cm away and the car moving at 10cm a minute.
The AI built a Discord bot in an evening that ingests an RPG PDF and stores it in a sqlite vector database for search.
He needed an app on his QNAP NAS that only existed as open source Linux code, with no NAS package. He gave the AI the model, the AI searched the web and gave him a downloadable package in about ten and a half minutes, and he installed it and it works.
The speakers on his Linux laptop had never worked. The AI fixed the audio, and he says he could not have done it himself. He now plans to check the code and try to send a patch to the Linux kernel.
Unite.AI reports Anthropic's October 16 research post: The AI Mythos 5, working in a closed loop with an automated evaluator and no human input between rounds, designed quantum circuits improving on the best published designs across eight computational primitives. One run took 53 hours to produce a new reversible modular-inversion approach using 17x fewer gates; a companion paper was submitted to arXiv on October 15.
AI through tscircuit and OpenAI models designed and autorouted about 15 ordered RP2040 and ESP32 PCB designs; known mistakes include improperly rotated connectors and routing loops.
Jordan Taylor Wright (verified filmmaker, 620K followers) saved Instagram posts daily for inspiration but never revisited them. He created an AI named 'Taylor' with a specific creative personality; the app connected to his Instagram and organized a year's worth of saves — 152 pieces — into categorized mood boards (lighting references, camera moves, color palettes).
A home-lab owner reported on X (via WhatAgentsCanDo, Oct 19) that, still in bed, he sent the AI a link to a post containing a setup recipe for a new DeepSeek model (DeepSeek-V4.1-Flash-1017, EXL3) across two NVIDIA DGX Spark machines and asked it to do the setup. The AI was not installed on the Sparks, but they shared a network with its computer, and everything was ready to test by the time he got up; he says he does not know exactly how the bot reached the machines.
The AI built and optimized his local decision engine for his private image organizer, reaching just under 85 percent accuracy across 500 validation prompts with sub-second responses.
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.