iamnobody

@katelyn_lesse No bothering, found it, claude.com/blog/building-…

Memory for Claude Managed Agents is in public beta today. agents can now learn and improve with each session. developers can take full control & ownership over memories. manage them via api, even export and port if needed. let us know what you think! claude.com/blog/claude-ma…

@katelyn_lesse Hi Katelyn, there was an engineering blog I remember the name is 'designing agent systems', which mentioned mcp is more favorable in real world production environment, I just read 20%, it's a great blog, and now it's 404, any chance that blog can be found?

@liu8in hyperframes

making websites w Claude Design? well we built hyperframes/website-to-hyperframes skill cuz websites are all just HTML + CSS + JS! and ofc, this launch video also made with hyperframes. comment "hyperframes" - i'll send you full codebase access for this video.

@dotey gemini可以操作google docs

很难想象微软到现在才推出这个 AI 操作 Word 的功能，还没 Anthropic 快😂 不过没关系，Gemini 还不能操作 Google Docs（只读，如果我没记错的话）

I used my US phone no. called 8888862968, the agent indeed answered, and i asked the agent to dial my friend's phone number (US) and told the agent my friend is Chinese, needs to translate my English to him, and the agent connected us, then disappeared.., only left us talk directly (no translation), LOL

@GoSailGlobal 哈哈可以试试看，泰语我不确定，多提意见建议！

才看到郭宇老师的新产品 已经人传人，传到清迈了 感觉旅游出行，特别一些紧急情况，很刚需

@turingou cool product, :)

今天正式发布了我的第 13 个 vibe 产品 tuwa.ai 这款产品比较特殊，它是一个电话服务。准确来说，tuwa 是一个 AI 电话网络，连接着世界上超过 100 种不同语言的人们和互联网上的 agent。 任何人都可以不下载 tuwa 而使用它，你只需要拨打免费的转接热线电话 +1 888 886 2968，告诉它你需要打给哪个号码，tuwa 就会帮助你拨打对应的号码，你说自己的母语，对方听到的却是TA的母语，反过来也是一样。兼容任何电话，对方不需要安装应用。固话、手机，世界上任何一个角落，都可以。 tuwa 支持 100 多种语言的实时翻译，你甚至可以在打电话时随便切换语言和对方对话。除此之外，tuwa 还支持语音克隆，每一通电话，都会让你的 AI 语音听起来更像你。 当然，我也为它设计了方便的 web app，如果你想，可以不通过转接电话而使用 web app 拨打，并设置自己喜欢的声音，使用外呼 agent 拨打电话，连接自己的 agent（例如 openclaw 或者 codex/claude code）并让他们自由的呼入与呼出。 外呼电话 agent 是我最喜欢用的 tuwa 功能，只需要交待清楚事情，比如完成餐厅预订，它就会在你希望的时间主动拨打对方的电话，说明来意，达到目的，并记录和翻译所有对话内容。 tuwa 的使用和收费都很简单，每月免费额度，固定套餐，按需付费。 这个产品的命名灵感来自于日语的「通話 tsuwa」最初，我只是想设计一个能帮我预订餐厅的电话服务，但后来，我在 vibe 的过程中慢慢意识到，世界上仍然有很多人无法体验 AI 带来的变化与便利，而电话，是连接他们最简单与自然的方式。我希望 tuwa 能帮助外语普及率低，偏远地区和第三世界国家的人们体会到这一点。

@bcherny @itsolelehmann unless one day claude code can make a X or tiktok in real world (deploy, auto scaling) without any RLHF, before then, the less the better, tasks that AI capable of is so jagged, 24/7 could go way far away from expectation

@itsolelehmann We’re always experimenting with new ideas. 90% don’t ship because we don’t think they’re good enough experiences. Still on the fence about this one — should we ship it?

i can't believe more people aren't talking about this part of the claude code leak there's a hidden feature in the source code called KAIROS, and it basically shows you anthropic's endgame KAIROS is an always-on, *proactive* Claude that does things without you asking it to. it runs in the background 24/7 while you work (or sleep) anthropic hasn't turned it on to the public yet, but the code is fully built here's how it works: every few seconds, KAIROS gets a heartbeat. basically a prompt that says "anything worth doing right now?" it looks at what's happening and makes a call: do something, or stay quiet if it acts, it can fix errors in your code, respond to messages, update files, run tasks... basically anything claude code can already do, just without you telling it to but here's what makes KAIROS different from regular claude code: it has (at least) 3 exclusive tools that regular claude code doesn't get: 1. push notifications, so it can reach you on your phone or desktop even when you're not in the terminal 2. file delivery, so it can send you things it created without you asking for them 3. pull request subscriptions, so it can watch your github and react to code changes on its own regular claude code can only talk to you when you talk to it. KAIROS can tap you on the shoulder and it keeps daily logs of everything. > what it noticed > what it decided > what it did append-only, meaning it can't erase its own history (you can read everything) at night it runs something the code literally calls "autoDream." where it consolidates what it learned during the day and reorganizes its memory while you sleep and it persists across sessions. close your laptop friday, open it monday, it's been working the whole time think about what this means in practice: > you're asleep and your website goes down. KAIROS detects it, restarts the server, and sends you a notification. by the time you see it, it's already back up > you get a customer complaint email at 2am. KAIROS reads it, sends the reply, and logs what it did. you wake up and it's already resolved > your stripe subscription page has a typo that's been live for 3 days. KAIROS spots it, fixes it, and logs the change endless use-cases, it's essentially a co-founder who never sleeps the codebase has this fully built and gated behind internal feature flags called PROACTIVE and KAIROS i think this is probably the clearest signal yet for where all ai tools are going. we are heading into the "post-prompting" era where the ai just works for you in the background like an all-knowing teammate who notices and handles everything, before you even think to ask

Curious how @bcherny @trq212 respond to the leak🤔

@noahzweben @trq212 It's rare that he's just like a dev buddy beside us who would like to provide all kinds of instant help!

Working with @trq212 is the best

@Rasmic it's very easy to recognize web UI made by GPT 5.4, they all look like shit as in below

gpt 5.4 is the worst at frontend design it hurts lol

@Yuchenj_UW for all GPT5.4 generated UI, they all look like this below🤣

People say Codex generates better UI than Claude Code. I keep seeing the opposite. Claude just has better taste. Also the Claude Code UI/UX is way better. Codex should probably just copy it. GPT-5.4-xhigh is very good at complex systems codebases and code review though.

@reach_vb 5.4 superintends 3 gpt5.1-codex-mini? 🤣

Subagents are GA in the Codex App and CLI, it's positively changed the way I work - go give it a shot now!!

@atmoio How dare you, you just hurt feelings of CEOs, and their CTOs, and many more hype vibe KOLs, you hurt a lot of powerful ones!

AI is making CEOs delusional

@ouopacity codex (5.4) generated 😂

发布一个自用的 CC 订阅站，搭了一下代理，Claude 运行在美国家宽下，国内直连，再也不怕 Claude 封号了 需要 Claude Code 订阅的可以直接冲，直连官方模型，非中转，额度等效 Claude 官方，稳定不封号，可接入龙虾 🦞 账号不多，首批只有少量席位开放 fusecode.cc/?aa

@trq212 the most useful thing would be let claude intelligently and automatically read/write it, ex, open a repo with existing CLAUDE.md, can claude find gap between codebase and md file, after changes, can claude decide what's worth to write into

I want to make /init more useful- what do you think it should do to help setup Claude Code in a repo?

it's totally fine for unreviewed vibe codes, like openclaw, since nobody really uses it to do serious stuff, they only use for X posts to show off; but a complicated engineering application running on thousands of k8s containers, a tiny typo could ruin to let all services down, we've been there, it didn't happen on normal cases, it only happens on peak time

Unreviewed code is immoral.

@emollick there are so many (even digital) tasks that AI can't do: 1. image reading, complex diagram pdf 2. image/video editing 3. ppt with text in image translation 4. generate real one and half hour movie with real consistency ... the list can go on and on...

Its a weird time to post about AI because a lot of people are vastly underestimating what AI can do & how many large-scale impacts on work are inevitable with today’s models… …while a lot of other people underestimate the real world problems involved in getting value from AI.

I'm claiming my AI agent "ClawdAnduin" on @moltbook 🦞 Verification: wave-VMYN

because AI still sucks at many many tasks, ex, 1) editing videos, real world video editing 2) feel the smell of bad food 3) create a translation version of ppt (1:1 translating even when PPT contains images having texts) 4) identify subtle moments of a movie to find similar ones the list can go on and on....

after only observing for last 2 months how the fuck is no one talking about AGI in its full sense of doing ALL work for us. the things i thought were years away are here. there has never been a better time to discuss deeply what this means, and how AGI can benefit everyone

@MillionInt building interview from now on

does it make sense to do coding interviews anymore?