Algo Trading, Strat, Macro. Math PhD (Erdős 3). Hon Reader@UCL. Created MSc Algo Trading (‘17-now), & teaching online. Too much street-side experience to list.
London, UK
Joined April 2009
- Tweets78.8K
- Following8.3K
- Followers7.5K
- Likes80.2K
Why give airtime to that low IQ idiot. What he knows is a small subset of other commentators. Yglesias is a severely retarded man.
His moral clarity is exceptional. It is true. It does not detract from the misery of those who escaped the Holocaust, but it is not the same as being a Holocaust survivor.
Objectively Random retweeted
全世界包括清北的编程课程都应该向斯坦福的CS146S Fall 2026看齐！这门课相比一年前的Fall 2025已经砍掉了85%的内容，可以知道编程领域进展有多恐怖了，如果还停留在教科书十年不需更新的幻觉中，可想而知会多落后了！
两版课纲我逐周对了一遍：
第 1 周：教你怎么写 prompt → 200 行代码手搓一个 Claude Code
期末项目占分：80% → 50%
给真实开源项目提 PR：0 → 30%
去年还单独讲一周的 AI IDE、Warp 终端、一句 prompt 生成整个应用，今年全部都拿掉。
取而代之的是 Agent Skills、CLAUDE.md / AGENTS.md / hooks、agent-ready 代码库、后台并行 agent、MCP portal 和模型路由，最后一周单独讲软件工厂。
刚入门的话，我建议从第 1 周跟起：
9/22 那节 200 行手写 coding agent，slides 和完整代码都公开了，最核心的是 agent 原型—— read / write / edit / bash 四个工具加一个循环
9/24 拆主流 coding agent 的 system prompt，原始 prompt 文件也已放出来
如果已经天天跟 coding agent 打交道，前几周大概率扫一眼就行，我会重点盯这几块：
- 嘉宾：SpaceX SI负责人 @leerob、Claude Code之父 @bcherny 已经分享完，10/22 轮到 Factory 的 CTO @EnoReyes，后面还有 Cognition、Semgrep、Cloudflare、Replit 等的 @amasad 和 @eladgil
- 第 9–10 周：大团队怎么落地，MCP portal、LLM gateway、模型路由、成本优化
- 前五周的阅读清单，凑起来就是一份 2026 版 agent 工程书单
30% 的开源分也很有意思：Vercel、Semgrep、OpenHands、Unsloth 等 15 个开源项目给学生做指导，PR 直接往生产级代码库里提，这在大学课程里还是头一回。
课程材料陆续公开，不是斯坦福学生也能跟：themodernsoftware.dev
后面几周的嘉宾分享我会边看边拆，关注我，第一时间同步最佳实践🫡
I’m excited to finally announce the newest edition my Stanford course 𝗧𝗵𝗲 𝗠𝗼𝗱𝗲𝗿𝗻 𝗦𝗼𝗳𝘁𝘄𝗮𝗿𝗲 𝗗𝗲𝘃𝗲𝗹𝗼𝗽𝗲𝗿. It has been 9 months in the making.
Last November, with the release of Claude Opus 4.5, coding agents experienced a step function improvement in capability. We all felt it. The LLMs were more powerful, could reason for longer, solve harder tasks.
This year’s iteration of my course reflects the 2026 metamorphosis of software engineering.
My core belief is simple: AI-native developers of the LLM era are going to become the most important members of any software organization. I have designed my course to train this next generation of engineers.
𝗪𝗵𝗮𝘁’𝘀 𝗱𝗶𝗳𝗳𝗲𝗿𝗲𝗻𝘁 𝘁𝗵𝗶𝘀 𝘁𝗶𝗺𝗲 𝗮𝗿𝗼𝘂𝗻𝗱
First, 85% of my Fall 2025 class material is being thrown out. The Fall 2026 syllabus reflects the core capabilities AI-native engineers must have: agent skills, advanced context engineering, MCP portals, agent-ready codebase principles, agentic code review, security, parallelizing background agents, software factories, and more.
Second, I am going to teach my students how to have software taste. Every student will be required to ship pull requests to production-grade, real-world codebases. The course is collaborating with the top open-source AI repos who will offer support and mentorship to students on how to meaningfully contribute to their projects.
This has never been done before in any university course so I am incredibly grateful to our OSS Partners: @browserbase, @HeyGen, @CopilotKit, @semgrep, @OpenHandsDev, @milvusio, @marimo_io, Pi, @crewAIInc, @warpdotdev, @vercel, @cmux, @arizeai, @UnslothAI, and @anyscalecompute.
𝗪𝗵𝗮𝘁’𝘀 𝘀𝘁𝗮𝘆𝗶𝗻𝗴 𝘁𝗵𝗲 𝘀𝗮𝗺𝗲
I’m fortunate to again have AI software engineering leaders and founders as guest speakers to share their learnings from building top coding agent products. Thank you to @leerob from @cursor_ai, @bcherny of @claudeai code, @EnoReyes of @FactoryAI, @silasalberti of @cognition, @0xine of @semgrep, Rajesh Bhatia of @Cloudflare , @amasad of @Replit, and @eladgil.
All resources will be available online. All classes will be available to the public.
9/22 on Stanford campus. See you in class.
themodernsoftware.dev/
Objectively Random retweeted
I’m excited to finally announce the newest edition my Stanford course 𝗧𝗵𝗲 𝗠𝗼𝗱𝗲𝗿𝗻 𝗦𝗼𝗳𝘁𝘄𝗮𝗿𝗲 𝗗𝗲𝘃𝗲𝗹𝗼𝗽𝗲𝗿. It has been 9 months in the making.
Last November, with the release of Claude Opus 4.5, coding agents experienced a step function improvement in capability. We all felt it. The LLMs were more powerful, could reason for longer, solve harder tasks.
This year’s iteration of my course reflects the 2026 metamorphosis of software engineering.
My core belief is simple: AI-native developers of the LLM era are going to become the most important members of any software organization. I have designed my course to train this next generation of engineers.
𝗪𝗵𝗮𝘁’𝘀 𝗱𝗶𝗳𝗳𝗲𝗿𝗲𝗻𝘁 𝘁𝗵𝗶𝘀 𝘁𝗶𝗺𝗲 𝗮𝗿𝗼𝘂𝗻𝗱
First, 85% of my Fall 2025 class material is being thrown out. The Fall 2026 syllabus reflects the core capabilities AI-native engineers must have: agent skills, advanced context engineering, MCP portals, agent-ready codebase principles, agentic code review, security, parallelizing background agents, software factories, and more.
Second, I am going to teach my students how to have software taste. Every student will be required to ship pull requests to production-grade, real-world codebases. The course is collaborating with the top open-source AI repos who will offer support and mentorship to students on how to meaningfully contribute to their projects.
This has never been done before in any university course so I am incredibly grateful to our OSS Partners: @browserbase, @HeyGen, @CopilotKit, @semgrep, @OpenHandsDev, @milvusio, @marimo_io, Pi, @crewAIInc, @warpdotdev, @vercel, @cmux, @arizeai, @UnslothAI, and @anyscalecompute.
𝗪𝗵𝗮𝘁’𝘀 𝘀𝘁𝗮𝘆𝗶𝗻𝗴 𝘁𝗵𝗲 𝘀𝗮𝗺𝗲
I’m fortunate to again have AI software engineering leaders and founders as guest speakers to share their learnings from building top coding agent products. Thank you to @leerob from @cursor_ai, @bcherny of @claudeai code, @EnoReyes of @FactoryAI, @silasalberti of @cognition, @0xine of @semgrep, Rajesh Bhatia of @Cloudflare , @amasad of @Replit, and @eladgil.
All resources will be available online. All classes will be available to the public.
9/22 on Stanford campus. See you in class.
themodernsoftware.dev/
Objectively Random retweeted
New Essay
I spent many weekends this year on automated theorem proving with AI + Lean, only for the Bitter Lesson to reign supreme. Some reflections on the OpenAI Math Repo and the short, one-summer window of human-AI collaboration in mathematics.
essays.johnloeber.com/p/34-b…
Objectively Random retweeted
New on our Frontier Red Team blog: using Claude to formally verify software at large scale.
Formal verification can prove that code is free of certain bugs, but historically it has only been feasible for small codebases (or required a lot of expert time).
With AI models, however, this is beginning to change. In collaboration with @theoremlabs, we used Claude to prove security properties in over one hundred real-world open-source libraries over the span of a few weeks.
Read our research in full here: anthropic.com/research/forma…
Objectively Random retweeted
On Lindsey Graham.
Stay with me on this.
The worst kept secret in Washington DC came out today because 3 brave men came out and told the truth and it doesn’t matter if it is after he’s dead.
Brave?
Yes brave.
Why are they brave?
They had to go on camera and tell intimate details of their personal lives.
To the world.
That is extremely difficult to do. Yes it’s brave. And Stressful. They walked into the public square opening themselves to the opinions of the entire world.
No matter your opinion on the issue, they took a courageous leap in coming forward.
If you are critical, would you go on camera and tell your secrets to the entire world?
I would bet a lot of money you wouldn’t.
Ok but there is more to this.
Because this is the world I saw up close in Washington.
I saw behind the curtain and saw ALL the wizards.
Everyone in the highest levels of power in Washington DC are keeping secrets that would destroy their entire lives.
And that’s exactly why they have power.
Not because they are so smart or qualified or the best choice for leadership.
I mean really, if you think our best and brightest are in charge then do some critical thinking and look at the pathetic state of our debt, our dollar, and how our government is run, and the outcome that affects your life.
Their secrets are the only reason these people rise to the top.
Because their secrets are not secrets.
We all knew Lindsey Graham was gay. Everyone did.
Their secrets are levers of control. Keeps them in line. Makes them a team player. And will make them fight the hardest on issues that don’t make sense in order to keep their secrets just that. Secrets.
So everyone is arguing about how this is old news, why now? Or arguing about being gay, and gay issues. Or arguing because that’s what everyone does on the internet, just argue with each other and with bots about issues they don’t have the real truth on, but throw out their strongest opinion anyways.
I’m here to tell you the granular issue you need to dial in on and block out the rest of the noise.
Three gay men just bravely pulled back the curtain and tried to teach you all the most important lesson America needs to learn.
It’s this.
America’s most powerful leaders keep secrets that comprise them in the most dangerous ways for the security, safety, and prosperity of our nation.
That’s the whole point.
I tried my hardest to fight the entire system on both sides of the aisle and couldn’t beat it because YOU ALL keep voting these people back into office and back into power.
By the way, I always cosponsored the Term Limit bill and no matter who is in charge, Pelosi, McCarthy, or Johnson, Dem or Rep, that bill never ever got even close to getting a vote on the House floor.
So the question for you, The American People, is are you able to learn the lesson and if so, what exactly are YOU going to do about it?
Ask yourself, what are you voting for? Donating to? Volunteering for?
And are you really satisfied with the outcome election cycle after election cycle?
Objectively Random retweeted
The next time Nick Clegg in invited onto BBC to pontificate as an independent tech expert…please note that here he is, named in the $27bn New Mexico judgment as one of execs engaged in a systematic cover-up of Facebook’s deceptive consumer practices.
Returns are returns to wealth. The stock market may just be incidental. Hugh is thinking in western DM terms.
Objectively Random retweeted
Understanding the Inner Workings of a State-Space Foundation Model for Limit Order Books arxiv.org/abs/2610.28753
AI in the hands of the uncritical turns out very differently to AI in the hands of someone who knows, though.
Objectively Random retweeted
It's happening. We're finally moving from the artisanal, bespoke world of chalkboard mathematics to large-scale industrialization in the mathematical sciences.
current.fas.harvard.edu/stor…
Objectively Random retweeted
Explosive story many years in the making, now out in Vanity Fair by Matthew Cole (who will be on Breaking Points next week, and publishing soon in @DropSiteNews):
Saudi Arabia and the UAE set up multi-billion slush funds for Jared Kushner to use.
Iran-Contra 2 here (except with Iran-Contra, at least the US got some hostages released. Not clear what we got out of this one.)
Yes and no. The people have turned but the elite are still owned.
The president was going to sign away indo sovereignty to the US (military flyovers automatically approved) but then the protests started.
Objectively Random retweeted
This is how the media works.
The Hollywood Reporter has been chasing us for comment, so they know exactly what our Ali G campaign is about. Instead of putting that to Sacha Baron Cohen, they handed him a different premise: that calls for cinemas not to screen the film are really about his support for Israel.
It’s the same link the Daily Mail and The Telegraph have tried to draw. It wasn’t true there, and it isn’t true now.
But it did its job. Cohen never had to answer whether Ali G is racist. He got to talk about Jews being targeted instead.
See how they're all working in conjunction? Legacy media sets up the talking point, then hands it over for the subject to run with.
So back to the question nobody in this exchange wants to touch. Ali G trades on a caricature of Black British and Jamaican culture. That’s our complaint. It always has been, and thousands of people backed it.
