A year ago I'd never opened a terminal. This week I published my first open-source CLI to npm.
No CS degree, full-time day job, built mostly between 8pm and 10pm (when the kids are in bed)
@gregisenberg Yeh they def are but I tried a week without codex and Claude an used open source models instead and I missed the frontier guys wayyyy too much. Give it a few more months
Claude Security scans now run on Claude Mythos 5, available today in public beta for all Claude Enterprise customers.
Put our most capable security model to work on your codebase, no separate model access needed.
On August 17, GitHub experienced a significant outage that disrupted developers and organizations around the world. If you were trying to ship software that day, we let you down.
I posted in March and April about the steps we’re taking to make GitHub more reliable. The work is still underway. Here is an update:
https://t.co/9KaQy7O4CG
@petergyang Agreed. I’ve been testing this theory with a ‘do something smart’ agent to see how results differ. They have all the context and tools they need to do the right things and I am the bottleneck 90% of the time
This blog post was written for those who may be curious to know what an agent harness is, but don’t, and might not know where to start. Read the full post here
https://t.co/pUnUnQvEHJ
TrueFoundry's new AI agent harness undercuts Claude by up to 75%.
Cheaper agents mean more experiments, more failures and more weird little tools.. San Fran is playin the long game.
F me, I haven't used Claude design for a few weeks and I just tried it with Fable and it fxckin knocks design work out of the park. Well impressed. Just wish it wasn't so token hungry.