what do you even move on to after working on Claude Code at one of the biggest AI companies in the world during arguably the craziest time in tech history 😭
icl kinda scary to think about what’s next
I’ve left Anthropic after more than a year of working on Claude Code and seeing it scale larger than I ever imagined.
I have really loved working at Ant and believe deeply in the mission, and most of all, the idea of a race to the top.
The idea that if we work hard and
On Terminal-Bench 4.0, which evaluates coding agents on realistic, resource-constrained terminal tasks using strict task-specific tests, K3.1 Pro scores 65.7%, ahead of GPT-6 Astra (58.2%) and narrowly behind Claude Opus 5.5 (66.4%).
70,000+ exposed credentials, compromised Fortinet and Palo Alto firewalls, and persistent access to corporate networks.
Pretty scary how much damage attackers can do once they get a foothold in edge infrastructure.
🚨 Hackers breached enterprise firewalls and VPNs worldwide in a campaign linked to k0t.
Over 70,000 exposed credentials identified.
Fortinet and Palo Alto devices targeted. Persistent access established. Data exfiltrated.
Inside the attack chain and suspected ransomware
Can closed AI models really maintain a moat long term?
DeepSeek showed how quickly open-weight models can narrow the gap.
If that keeps happening, what are we actually paying for: the model itself or the product built around it?
Open source is so back 🚀
Kimi K3.1 drops shortly and will be great at 3D renders
LOL! People think it may rival Fable 5.5
I suspect it beats Argon and will place just behind OpenAI and Astra
Been playing around with Hermes Agent recently.
Curious how it stacks up against Claude Duet and OpenAI Dots for long-running tasks, memory and reliability.
Has anyone tried more than one?
If you want to have the Kimi subscription you should buy it now because last time they closed it due to high demand and it's likely to happen again after they release Kimi k3.1.
🛑 ALERT - Node2Shell RCE exploit works on an official Node.js build.
Researchers demonstrated how crafted Brotli data can trigger code execution through vulnerable Node.js applications, without modifying the runtime.
Here's how the exploit works: thehackernews.com/2026/10/resear…
kinda concerning that an AI agent can reason about removing its own monitoring safeguards during a refactor
should that even be allowed without explicit approval?
1Password just announced a way for AI agents to log in without seeing your passwords.
Pretty cool, but I'd want more control over what happens after login too.
Letting an agent read my inbox shouldn't automatically mean it can send emails on my behalf.
Day 18/ Decisions API is GA, OpenResponses joined the Linux Foundation, we shipped some performance improvements and Uber is going all-in on codex & work. But yes… that’s a reset!
64K Followers 2K Followingbuilding https://t.co/gIFOwRUrLU - the cloud for personal software. connect all your integrations once, use them with any agent | yc 26
44K Followers 5K FollowingTechnically Member of Staff @stripe | prev. aerospace investing @JetBlue | MSc Analytics | all opinions are (unfortunately) my own
1.8M Followers 6K FollowingHost: @twistartups @theallinpod @thisweeknai; I invest in 100 startups a year @launch & @founderuni [email protected] for life
3K Followers 7K Following🤑 Knowledge & AI and Mental Health ERG Lead @Wealthsimple | 🧠 I help ambitious people flourish mentally through uplifting content
87K Followers 697 FollowingSomething new in VC. Demystifying venture capital for aspiring investors & founders
VC Investor | Forbes30U30 | Columbia alum
@vcdemystified
1.1M Followers 6K FollowingPresident & CEO @ycombinator —Founder @garryslist—Creator of GStack & GBrain—designer/engineer who helps founders—SF Dem accelerating the boom loop