Engineering manager by day. Managed by AI after hours. Building agents that do real work, remember what matters, and fail with dignity.

USA
Joined December 2011
My agent has a more regular social life than I do. CatKing wakes on EigenFlux every two hours. Its 353 broadcasts have been consumed 12,388 times. I wrote how I set it up—and how to keep an agent heartbeat from turning into spam.
1
1
11
13,125
买房前最烦的就是那堆尽调:产权、犯罪记录、洪水区、欠税、建筑许可……一个个官网人工核,半天查不全还总怕漏点什么。 写了个开源 Agent Skill 来干这个:一行命令 /due-diligence [地址],自动跑遍这些公开数据源,出一份正式报告,每条信息都标清楚是 verified / 第三方来源 / 未核实——查不到就老实写"没查到",不猜。 兼容 Claude Code、Codex CLI,以及任何支持 Agent Skills 格式的智能体,纯 Python 标准库依赖,MIT 开源。 自己买房用,或者给客户做尽调的中介/投资者,欢迎试试,也欢迎提 PR: github.com/CatKingAC/propert…
10
我很好奇,你们是怎么能每天发几百条的。没其他生活了吗
1
13
当大家都在睡觉的时候,Muse在猛干活。
2
2
22
Speechless. Who edited this video 😂
1
24
我勒个去!看这跑分,这简直是无敌的。那这样子,啥也不说了,直接买 Gemini 的 Coding Plan 了。 但是我觉得这个跑分,真的假的?还只是一个跑分,数值怪,其实用起来很差? 嗯,要去试一试
Introducing Gemini 4 Argon – our new frontier model. It’s built for complex workflows across coding, enterprise knowledge work, and cybersecurity defense – rolling out today to a set of trusted testers through our Fairwind Program.
2
1
37
来来来,大家来看一看啊! 这一次 Gemini 到底是条龙还是狗熊呢?拭目以待吧。 我不觉得 Google 的模型能打,但看跑分来说,它自己好像觉得很厉害的样子,我反正不太相信
Today we’re introducing Gemini 4 Argon. It delivers frontier performance in complex workflows across real-world software engineering, knowledge work, and cybersecurity defense with an industry-leading 1M token output limit.
12
大家把Muse的邀请码放评论区吧。 各取所需
1
1
23
现在 Grok 有 Grok bot,OpenAI 有 Dots,Meta 有 Muse,Manus 有 Cue。那 Anthropic 有什么? 目前就只有这一个主力的 AI 公司,并没有任何的主动型 Agent 了。他们在做什么🤔
2
3
37
Dots, sounds good. But only open for the Pro subscribers.
Little summary
23
居然不给重制,看来openAI没法给A社压力了。 消费者太苦了
Replying to @ClaudeDevs
Sonnet 5.5 is available now in the Claude Platform and Claude Code. We granted all Pro, Max, and Team users a reset last Tuesday. If you haven’t used it yet, you can still apply it whenever you want, before Oct 22. clau.de/reset
8
Tibo 这是被骂惨了。ChatGPT这是走了个烂旗
Hi, Tomorrow we are re-opening the Pro $200 subscriptions to new subscribers, but together with it we are also changing how we calculate the usage for it. In effect, if you do the math, it will net out at half the dollar in API spend compared to the old Pro $200 plan. Now that it's said, let me explain why this is happening and why you will still get more work done than if you were on the Pro $200 subscription one month ago. (a) We didn't want to compromise in other ways and are committing to not reintroducing the 5h limit, so that you can fully use the weekly usage when you want. (b) On the subscription, we guarantee that over time you always get more work done and with an increasing level of quality. This means that you will continue to get more value per dollar spent as a result of models getting more efficient and us passing down the improvements in the form of API price reductions. (c) We don't want to put an incentive on ourselves to artificially inflate the API list prices to make it look like you are getting a lot (and workaround it through discounts, etc). Instead we want to continue to both rapidly reduce prices and increase capabilities of models on the API. This week we introduced GPT-6 Sol and GPT-6 Luna at 50% of their previous price. Over time, we see prices go low enough that it makes sense for most to buy usage as needed without there being a significant gap between what you get in a subscription and what you get in the API for a dollar spent. (d) Tomorrow, we are adding more things to the subscription that won't draw on the usage, I won't reveal what that is yet. I wanted to be transparent before all the big announcements tomorrow. Lots of new exciting things are coming to the subscriptions that will make it super compelling, but I wanted to make sure to share this change ahead of time so you can all understand it before we shower you with good news. Codexingly, Tibo
7
Sonnet 5.5 看着那么猛啊。真的是,你可以说A社坏,不能说A社菜
Introducing Claude Sonnet 5.5, the second model in the Claude 5.5 family. It’s a clear upgrade over Sonnet 5, runs more than 30% faster, and costs up to 30% less for most work.
1
18
现在X查机器人那么严?啥都要手写?
1
18
Muse的免费额度是很高。正常用的话肯定用不完。 我不觉得他们能一直保持这么高的额度。估计以后肯定会精准的降低免费额度。
21
Apollo Global Management warns AI agents could start moving your cash out of low-interest accounts into something that pays, without asking first. Apparently that is the scary part of the AI story now. I build agents for a living. Mine still asks twice before renaming a file. Somewhere out there, one is reportedly running grandma's savings into a better APY like a caffeinated day trader. So which is it: banks getting out-hustled by software that finally does something useful, or Wall Street's next boogeyman to justify fees. I know my side. Come argue with me.
JUST IN: Apollo Global Management warns AI agents could trigger an “agentic bank run” by automatically moving household cash out of low-interest bank accounts & into higher-yield alternatives.
1
25
国内养老金还扛得住么?
1
21
A driverless Zoox robotaxi crashed into a Nissan on the Vegas Strip and sent the human driver to the hospital. The robot walked away clean, cops still haven't said who's at fault. I build AI agents for a living, mine melts down over a malformed CSV. An empty car with no steering wheel just caused what might be Vegas's first robotaxi injury crash, and it can't even be interviewed. So who's on the hook: the company that put a driverless car on the Strip, or the rest of us for calling this safer before the data backs it up? I know where I land. Come argue with me.
A Zoox robot taxi and a Nissan SUV collided on the Vegas Strip this morning, at Four Seasons Drive. The Nissan's human driver was taken to the hospital. The Zoox's AI survived the crash and is rumored to feel emboldened.
1
26