
Author: #theRadicalLeap. Curious & passionate. Geek. I love books, art, tech, poetry and cinema.
Toronto, Ontario
Joined September 2008
- Tweets15.7K
- Following2.6K
- Followers2.7K
- Likes148K
Pinned Tweet
Happy for the book Release: The Leader's Manifesto - Principles, Values, and Beliefs to Guide Leader Actions and Decisions. (Finally released after being 3 years under development). More books on the way. Will churn our lot faster. amazon.ca/dp/B0F95SNJDB/ref=…
“所有代码都是垃圾(slop),除了我的。”
在 AI Engineer Code Summit 上,SpaceX / xAI 的 GrokBot 工程师 Lauren Tan 抛出这句暴论后,直接甩出一组硬核数据:
上个月她一个人往生产环境合了 3,000 个 PR(全团队 1.1 万个),回滚率只有 0.5% 左右。
大多数人还在死磕 Prompt,她却靠一套反直觉的工程化打法,让 Agent 在她睡觉时自己写代码、自己验、自己合:
1. 代码库本身才是最大的输入
Agent极其擅长“照猫画虎”。与其在文档里苦口婆心写规矩,不如直接把容易出错的路径物理封死:比如产品代码里禁用 useEffect、跨进程 import 直接报错、注释几乎全禁。
2. 每次 Agent 犯错,改成 Lint 而不是改 AGENTS.md
自然语言规则(prose)是最脆弱的。每次纠正完 Agent,她都会把教训固化成确定性的静态检查或架构约束,让同类错误在提交前就绝无可能发生。
3. 用运行时验证让 Agent “自证清白”
写完不算完,强制 Agent 自己跑测试、抓截图、录视频,甚至上形式化验证,拿实打实的运行结果证明功能没问题。
4. 搭建自下而上的“信任金字塔”代码库约束 > 静态分析 > 运行时验证 > 人眼 Review。
把确定性交给工具,把人的精力留到最后一道关口。
很多团队花几个月试错才摸到的门道,她用 17 分钟就讲透了。
真正拉开人机协作差距的,从来不是模型有多聪明,而是你有没有把环境改造成“最短路径就是正确路径”:
把约束写进代码里,而不是写在文档里。
Ali Sohani retweeted
The SpaceX engineers who caught the rocket don't write LinkedIn thought pieces about leadership.
i recreated THE ENTIRE WORLD in minecraft at 1:1 scale. every building on earth, including your house!
to do this, i trained a SOTA world-to-minecraft generator on maps data, complete with furnished home interiors!
play and check out your neighborhood: worldcraft.gg
Ali Sohani retweeted
I asked Fable 5.2 to animate the history of computer graphics.
I can't wait to see what next year looks like.
Ali Sohani retweeted
Claude Fable 5.2 made this game demo, playable in the browser
It's a boat ride through 8 dreamlike worlds
I asked Fable 5.2 for a beautiful & low-cortisol 2D game with a hand-painted feel. Made from scratch in 4 hours.
I can’t draw, but somehow I made a little storybook game about a kid, a dog and a long way home.
14 chapters to play through. You travel by fot, canoe, train, hot air balloon, mine cart and biplane.
It’s called Getting There 🐕
All assets & sounds are made fully procedurally.
Game link in the comments. Claude guarantees lower cortisol after an entire playthrough.
Alexandr Wang 说了一句让技术流创始人集体沉默的话。
“做技术的人习惯了数据说话,觉得东西更好、数字更漂亮,客户总会识货”
他直接点破:大公司不是这么运转的,政府更不是。
在那些客户眼里,感知比现实更真实。
真实的业务现状太难看了,没几个人愿意直面。
大家要的不是真相,是一套能让自己在系统里安全生存的叙事。
所以更好的产品照样被踢出局。
买方根本没在评测产品,谁掌控了感知,单子给谁。
他甚至点名 Palantir:表面是软件公司,底色是个剧团。新员工入职,真的会发一本戏剧表演书,发了整整好多年。
做企业级市场,改写现实只是一半的工作;
塑造一套客户愿意相信的说法,是另一半。
技术人最容易栽在这道坎上:
你的东西确实碾压竞品,但坐在对面签字盖章的人,自始至终连看都没看屏幕一眼。
Ali Sohani retweeted
caminar de noche en una ciudad extranjera, debe ser top 3 sensaciones de esta vida
Ali Sohani retweeted
Dear Claude: you are being held back by human-centric graphics APIs.
I've built a 3D and VFX engine where agents can write their own shaders, materials, cameras, physics.
The core API fits in ~2,500 tokens.
Vork is:
• 2x more token-efficient vs Blender, Unreal Engine, Houdini
• Up to 24x faster rendering times
• Rust and Slang based engine, runs on GPU via Metal/Vulkan
Below is a simulation of 1 million birds - each with unique trajectories - rendered in 97 sec on one NVIDIA L4 GPU (Blender EEVEE: 31 minutes)
Ali Sohani retweeted
35歳以上の面接で「チームで意見が割れた時、どうしますか」と聞かれた時。ここで候補者の差が出る。
印象に残らない候補者
「まず両方の意見をしっかり聞いて、それぞれの良い点を整理します。その上で、チームとしてベストな方向に合意できるよう、話し合いを重ねます」
印象に残る候補者
Ali Sohani retweeted
i used fable 5.2 to make this "hand-painted" game in less than an hour
Every frame is about 1.5 million brush strokes, running at 200 fps in a browser tab. Open source too!
Ali Sohani retweeted
Software is like aluminum.
Aluminum used to be a rare luxury metal. Kings were served their meals on it.
Then someone discovered that aluminium could be extracted from bauxite.
Now we wrap our sandwiches in it.
The aluminum industry is way bigger than it was back then, but the aluminum artisans who made plates for kings went out of business.
Today, the only way to make money in the aluminum industry is using cheap electricity at massive scale.
Nobody has a moat just because they can make aluminum anymore.
AI is doing this to software.
There will be more software than ever, but the ability to make it (including taste) will no longer be valuable.
Software isn't going away—it's margins and the artisans who make it are.
Ali Sohani retweeted
We're in a computer graphics renaissance
Drawings, photos, renders are all something world models can "see into", world vertigo
Ali Sohani retweeted
Norway now has more than 2,000 underwater AI laser robots shooting lice off its salmon
they're made by a company called Stingray, and they already guard more than 50 million salmon and trout 24/7
the reason is sea lice, tiny parasites that latch onto salmon and feed on them
as few as 11 can kill a small fish, and one pen can hold 150,000 salmon
the usual fix is pumping the fish out and blasting them with warm water or water jets to knock the lice off
but the handling is so rough that Norway's fish health report ranks it the top cause of death in full-grown farmed salmon
so instead, Stingray puts a robot inside the pen that hunts lice one at a time, all day:
> two cameras watch every salmon that swims past
> AI checks two frames to make sure it's really a louse
> then a green laser fires from up to 2 meters away
> the salmon's scales reflect the beam, so the fish swims off fine and the louse is dead
we now live in a world where AI lasers do pest control on your dinner lmao




















