Keep on to blur preview images; turn off to show them clearly
RT @marclou: I built a marketplace to buy/sell startups: 📋 Founders list startups with open metrics 👀 Buyers make private offers 🤝 TrustMR…
💻 https://t.co/Y30jsaHwz9 $30K/m ⚡️ https://t.co/vatLDmi9UG $21K/m 📈 https://t.co/3EDxln5mdi $17K/m ⭐️ https://t.co/MZc8tG9xWi $17K/m 🍜 https://t.co/r07EpGSYJ2 $1K/m 🧬 https://t.co/SfrVXVtmdA $0/m 🧾 https://t.co/7olaOzV8Xd $0/m +20 https://t.co/4zCWHGJp1S
RT @AmandaAskell: I just want to confirm that this is based on a real document and we did train Claude on it, including in SL. It's somethi…
Claude is an AI assistant built by @anthropicai to be safe, accurate, and secure. Talk to Claude on https://t.co/ZhTwG8dz3D or download the app.
closing in on 100,000 downloads to my 1.5 month old app @lovelee_app :) funny because i thought i was CRAZY setting a $5k/month goal for a brand new app by end of year (8 week challenge) but guess i should trust myself more
curious guy creating things @ https://t.co/HXWladhJaA - up and coming wife guy
the beauty of Claude Skills is really in that it’s an amazing distribution mechanism for “agent capabilities as an atomic unit”, way better than what we had before Yeah I like the context management stuff, we all hate tool bloat. But I LOVE that the new convos become (oversimplified): - “oh that important business/dev thing you need to do? Add our Skill to the agent” - “We have a huge repository of pre-made skills and here are some other great ones from the community! You want something custom? Here’s our skill builder, it’ll make you a nice folder” Skills define a unit of capability unit for your agent harness much better than tools/MCP attachments they sort of move us along this curve of the deep user desire of “I want an outcome, not a tool”. It’s not an outcome exactly but it’s closer + I think it helps builders think about customer business goals which is what really matters, like “how can I package this agent ability to help my customer solve their problems” Not “ooo cool tool, imma build it” Claude Skills is the prototypical type of feature that attracts engineering scorn in the form of “it’s nothing special” or “it’s not real engineering” first of all it’s a nice but of context engineering at the very least, but also it’s a good abstraction. of course the real work comes in when making sure a set of skills actually works well together with your agent, because harness engineering is inevitable
building agents and harnesses, prev @awscloud, phd cs @ temple
Claude Code 在国内使用的最新方法？ DeepSeek 昨天推出了 v3.2，特别优化了推理和 Agent 支持。 重点是 DeepSeek-V3.2 的思考模式增加了对 Claude Code 的支持。 可以通过将模型名改为 deepseek-reasoner，或在 Claude Code CLI 中按 Tab 键开启思考模式进行使用。
DeepSeek V3.2 正式版：强化 Agent 能力，融入思考推理 https://t.co/SiGGyFUFNp
Imagine if this is another Kimi moment and they actually legit surpass DeepSeek’s current version …hard, yeah. But can’t we hope?
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