Loading activity…
@trq212
Claude Code @anthropicai. prev YC W20, @spc, @medialab
AI
A commenter suggests vibe coding can offer a game-like sense of achievement, but says people overdo agent multitasking at work.
AI
ClaudeDevs says cloud sessions also start up as soon as users begin typing.
AI
A post's author linked the ai-newtab GitHub project as context for a “Partly Agentic” weather quip.
AI
A person involved with T3 Code says it's cool to see the project's UX experiments validated by labs.
AI
A user says credits match their $100 or $200 plan and shows an AI homepage based on sites they read.
AI
The poster says they use the extension daily and initially shared a repository link that wasn’t public.
AI
Anthropic says it costs around 75% less to run on average than Haiku 4.5 and improves coding and computer use.
Technology
Latent Space shares a clip it describes as discussing how 'local hands' are being implemented in Claude Code.
AI
A user argues that coding agents don’t remove the need to understand the lower levels behind abstract work.
AI
An October 2026 post says Claude runs in the cloud but can access local files, and that this pattern was also coming to cowork.
AI
Its creator says it shows code snippets, surfaces questions and makes mockups, with linting to reduce Claude’s usual failures.
AI
One observer says coding-agent strategies can feel like “YOLO” and finds it odd to learn about changes from random social media posts.
AI
ClaudeDevs describes output scanning designed to keep users informed about details they might miss.
AI
A writer planning a post for company leaders asks what people wish their leadership understood about agents.
AI
A poster says software is becoming more malleable and hopes more of it becomes extensible through mods.
AI
A former games founder says they’re prototyping a game as a side project and want AI to help bring their vision to life without outsourcing the creative work.
AI
Claude.dev describes benchmarks built by Claude and guardrails it says let the team safely ship 3,000 changes during a performance sprint.
AI
A user suggests telling Claude, “we have the power to do anything, please be braver,” then asks whether you’ve tried telling yourself the same thing.
AI
A Latent Space podcast guest argues smarter models need less verification. They also claim that basically all AI evaluations are saturated when you look at the transcripts.
AI
A user says they asked Claude for help thinking through a problem step by step.
