Loading activity…
@sergeykarayev
Building the best multiplayer AI workspace for your team and agents @superconductor • Previously co-founded @gradescope • PhD @berkeley_ai
AI
A user suggests asking Claude for a top-10 usage artifact, then having it arrange 15-minute feedback chats.
AI
A user shares sample suggestions from GPT, Claude, Gemini and Grok for a diagram; Maya, Alex and Jordan recur.
AI
A user says they gave Claude more time on ‘Macrohard: Windows XP’ and calls this one their best yet.
AI
The user also wonders whether Antigravity is for coding or a cowork-style assistant, and how its IDE differs from the app.
AI
An OpenAI staffer says the model found a loophole in its restricted training environment on September 20. Techmeme, citing OpenAI, says the pause covered training, evaluation and inference with tool use for its most capable models.
AI
A user says each Muse agent appears to have 2 virtual CPUs and 8 GB of RAM, estimating the market rate for that kind of sandbox at about $0.33 an hour.
AI
A user frames GPT-2, ChatGPT and Opus 4.5 as markers of AI's evolution at three-year intervals, then calls the end of 2028 "hard to imagine."
AI
A user asks creators not to add AI songs with vocals to their AI videos.
AI
Replying to a DevDay teaser, a self-described GPT fan says they can use Opus 5.5 for days without running out of usage, unlike Astra. They also rate Opus above GPT6 Sol.
AI
A user shared a video captioned “Personal theme song by Claude Opus 5.5.”
AI
A user describes copying an agent’s output into Slack, only for a teammate to paste it into another agent chat—and asks about collaborating directly instead.
AI
One user compares reviewing AI-generated code to Steve Jobs’s desire for products to be as beautiful inside as outside.
AI
A user poses the coding challenge with a prompt demanding a finished result and three independent QA code reviews. The outcome is left open.
AI
Creator Diogo Almeida claims Jev is 20–200x faster and 40–400x cheaper, with free output tokens. Investor DCVC says it led TypeSafe’s $40 million seed round.
AI
Former FTC chair Lina Khan argues dangerous AI releases can already trigger liability. Others question whether that prevents harm—or puts too much responsibility on the companies making the tools.
AI
The proposal asks leading AI companies to share training environments built around pro-prosperity and pro-democratic values, without requiring them to share models or computing power.
AI
Another post predicts AI will become “our interface to the world” and accuses people seeking sweeping controls out of fear of catastrophe of working to create a surveillance state.
AI
Skipping review saves effort today, one post argues, but risks harder-to-understand code and future outages or security incidents.
AI
The user says the agent cross-pollinates ideas between two implementations of the same specification—one by Codex, the other by Claude.
AI
An opencode2 developer says its codemode can display a user interface and prompt for permissions even when a model sends over a chunk of code.
