Keep on to blur preview images; turn off to show them clearly
RT @shao__meng: 成本直降 90%，延迟缩减 85%：深度解读 Prompt Caching 技术原理 来自 @ngrokHQ 团队 @samwhoo 关于 Prompt Caching 的技术博客，Sam 从底层原理到厂商实测，非常清晰地揭示了这项技术如何成为…
邵猛，中年失业程序员 😂 专注 - Context Engineering, AI Agents. 分享 - AI papers, apps and OSS. ex Microsoft MVP 合作 - 私信/邮箱：shaomeng@outlook.com 📢 公众号/小红书: AI 启蒙小伙伴
There's a lot of talk about Xi breaking the Party convention of generational change. But if you look at it, what really broke is the American ability to elect leaders who were born after 1946 AD. Xi is 8 years older than Obama. Biden is 11 years older than *Xi*.
We're in a race. It's not USA vs China but humans and AGIs vs ape power centralization. @deepseek_ai stan #1, 2023–Deep Time «C’est la guerre.» ®1
i’m becoming more anti “one way door” patterns in Agents compaction is a one shot distillation of a huge piece of context that we can’t recover from Subagents are a one shot completion of a task that the orchestrator can’t get follow-up with we know one-shot things are often unreliable with agents but for these cases we’re cool with it?? 🧐 like Jeff said, it’s worth - use fs to offload and have some agentic search over it - maybe use a subagent for the search that can go back and forth with the orchestrator to get the important pieces - you rlly can’t know apriori what will and won’t be important as the trajectory evolves
agents, harnesses, and evals @LangChainAI, prev @awscloud, phd cs @ temple
when i talk to coding agents I talk like a king... "i have no patience for more failure" "this must be fixed or else" "you need to think harder on this" "fffffffffffffff" "Ugh!!!!!!!!"
Full-time Vibe Coder. Built the best slack well-being app and first iMessage video app. ex SDE @ twitter, tripadvisor, proof.
RT @Yoroomie: I'm finalizing our 2025 Recap & Year In Review for the Everything Marketplaces (@marketplaceshq) community. Who wants to he…
Founder of Everything Marketplaces (@marketplaceshq). Always working with & investing in marketplaces at https://t.co/lUdl8p1TzR
RT @HomelandGOP: "Sophisticated actors will attempt to use AI models to enable cyberattacks at an unprecedented scale." @AnthropicAI’s Dr.…
We're an AI safety and research company that builds reliable, interpretable, and steerable AI systems. Talk to our AI assistant @claudeai on https://t.co/FhDI3KQh0n.