AI工程师一枚,分享自己的日常学习与项目构建。 https://xitter.cf/t.co/UhzqKNgqVQ https://xitter.cf/t.co/n1pEaXJVYj

Joined June 2014
Pinned Tweet
👉 learnai.selfworks.ai Agent 会成为每个人的日常工具。理解它的工作方式,能帮你更好地使用它。 所以我做了 LearnAI:从开源 Agent 的真实源码出发,帮你看懂 Agent 是怎么工作的。 接下来还会建一个 LearnAI 微信交流群。感兴趣的朋友,注册后可以通过账户页面里的微信号添加我,一起学习和交流 Agent。
我自己做了一个Pi核心20道题,理解了这20道题,对agent就有了一个大概的了解了。
3
3
28
1,637
就以AI现在的发展速度,明年具身智能就到达一个明显的高度我也不会感到吃惊。
30
所以人类的智力不过如此。
1
1
78
某些人又要瘫坐眩晕了!
Exponential Time Hypothesis (ETH), a 27-year-old conjecture in complexity theory, just fell. On Oct 16, OpenAI and Anthropic models independently solved 3-SAT with the same 2^(n/√log n) bound. Not P = NP, but faster than anyone thought possible. We live in extraordinary times.
24
大佬们都觉得软件已死了。
2
2
244
Turning complicated hardware into simple, fun lessons | ESP32 edition Just like my Arduino app, this one shows you what the ESP32 can do, one circuit at a time. Made with Opus 5.5 learn-esp32.vercel.app
Hardware looks scary until you break it down into small steps. So I built an app that takes you through Arduino one circuit at a time. Made with Opus 5.5 learn-arduino-red.vercel.app
47
233
8
1,841
65,849
kayce001.github.io/sky-strik… ,做了一款射击类的小游戏,大家刷X累了,可以来玩
1
90
AI 大量写代码之后,传统代码审查方式已经跟不上了。 以前大家关心的是“有人看过代码了吗”,现在更该关心的是“这个改动可能出什么问题,该用自动化还是人工来拦住”。 人的时间应该留给真正需要判断的事:要不要做这个功能、架构方向对不对、高风险逻辑是否合理、出了问题谁负责。常规的错误检查,尽量在提交前就让 AI 自己跑测试、做验证,再用多层自动化(AI 审查、安全检查、测试等)尽量拦截。 更重要的是心态转变:不追求合并前绝对不出错,而是承认总会有漏网的,所以把“出问题后能快速关掉、快速回滚”做成最后防线。新功能先放在开关后面,先给内部人用,有问题直接关开关,不用重新发布;回滚尽量做到几分钟内完成。 衡量流程好不好,也不只看有没有人审批,而要看人类真正花了多少时间,以及回滚率和事故率有没有上升。如果回滚变多了,说明问题只是被推迟了,并没有真正解决。 核心就一句话:在 AI 写代码越来越快的时代,把人的注意力从“逐行检查代码”,转移到“控制风险 + 快速恢复”上。
7
4
441
早安!
2
136
这个有意思
This looks more fun than a robot duck... a $400 humanoid built around an old phone. UCLA’s RoMeLa team built PhoneBot from an Android phone and a modular body with 13 motorized joints. It can walk, turn and get up from a face-down position. The phone supplies the camera, motion sensors and processing power for vision and walking control. The team even ran the core software on a 2017 Honor 9. The team estimates the hardware cost at around $400, not including the phone. CAD files and the parts list are already online.
1
111
按照这个进度,周三就用完了
95
pi-mcp包
1
2
6
1,446
所以很多时候就看谁在发帖,内容不重要。
85
这哥们对自己的项目非常自信,T3code好过codex和claude
1
1
191
opus认为对我的影响很小
1
117
There’s a reason I tell everyone to skip Tailscale and setup WireGaurd directly Minimize the attacks surface areas, stop adding unnecessary links that are a supply chain disaster in the making
‼️ Researcher Erica Windisch skipped coordinated vulnerability disclosure and published 11 unpatched Tailscale flaws. In her lab, Tailscale SSH race conditions let one hacked device get root on other servers and add a rogue node.
87
55
10
929
95,924
操,这么快就轮到tailscale了
‼️ Tailscale SSH Flaws Could Let Hackers Gain Root Access Across Servers Details: cybersecuritynews.com/tailsc… A new security report warns that Tailscale SSH has flaws that could let a hacker move from one compromised device to root access on other servers. Tailscale SSH lets approved devices log in to each other over a private network. The report lists 11 findings in five groups. They cover Tailscale’s SSH server, its serve and app connector features, the DERP relay client, and headscale, an open-source server that manages Tailscale networks. The researcher tested Tailscale 1.104.2 and headscale 0.29.5 between October 22 and October 24, 2026. The most serious group involves Tailscale SSH. Its SSH server runs inside tailscaled, a background program with full root privileges on the machine. #cybersecuritynews
Readers added context they thought people might want to know
The flaws require an already-approved device in the Tailscale network with SSH rules permitting access/forwarding; they enable post-compromise lateral movement and escalation inside the tailnet, not unauthenticated remote access from outside. tailworm.com brandefense.io/security-news/… tailscale.com/security-bulle… x.com/mmaunder/statu…
78
我觉得我的follower群体是有品位的,用心写的浏览量才高。
2
5
395
自己做的agent已经初步可用了。
1
2
261