Ryan X. Charles

team monorepo needs today include * copy-on-write worktrees * full incremental graph builds/tests * shared caches and like most tooling was built for small repos! so we are not there yet

he’s burying the lede that they’ve already identified seven (7!!!) submerged precontact sites along the coasts of North America

End of an era. Was just watching the video that I shot of Jensen at GTC this spring. He was on top of the world. Had built one of America's best companies. Was talking to a hockey stadium with every seat filled. Took hours to talk with the press, answering every question. Now not just his personal life, but his business life is on fire. The WSJ and NYT are saying the board is meeting right now to fire him. One thing I learned at Microsoft is to keep your head when everything is going wrong. And this series of affairs has cost me some money, and probably will tomorrow. As it will many of you who are investors in NVIDIA. One thing about NVIDIA. It has a very strong bench of people who could run the company. You don't know that because Jensen was the only one who built a relationship with the market. He is so strongly tied to the company. Let's be honest. It's a huge loss for NVIDIA. Last week no one had any questions about the governance of the company. Now the knives are out and blood will flow in the board room. But NVIDIA will survive. It will see a better day again. Here's my expectations. That tomorrow two new committees will be announced: 1. An executive search committee. 2. A temporary executive committee. Knowing NVIDIA from a close seat. I'm very confident that it will continue doing well after this all gets cleaned up. The talent bench there is so deep. I'm not selling my stock.

This was always the plan of the Dems btw. You're seeing the shift now. They want to monopolize superintelligence and achieve reg capture with a few insiders.

The WSJ is reporting that the NVIDIA board is preparing to oust Jensen Huang as CEO, and is preparing to meet later today.

The PR multiplication that comes from multiple repos slows down development and review. Group related code together in git if work on one feature is likely to span multiple projects. You can have multiple monorepos, despite how silly that sounds, for unrelated groupings of projects.

"AI agents are the majority of Web traffic and generate most code pushed to production. Soon, they'll be buying most of what gets sold online."

look I love rust as much as the next guy but unless your code goes to mars, I don't really want to hear your takes on this

You’ll want to patch your nodejs deployments this week. Claude was able to discover, chain and successfully remotely exploit CVE-2026-123864, CVE-2026-123866 (another bug collision), and CVE-2026-123874. The first two are remote memory disclosures and the third can be leveraged for RCE, but requires the app to use stream.respondWithFile/FD

It's really wild working with abliterated models primarily. You ask it to do something, and it'll just fucking go. Wild. We've conditioned ourselves as an industry to toe step around AI and our AI overlords. No longer needs to be the case. The machine works for us now.

October in a nutshell: - PHP got hacked - Railway got hacked - OpenCode got hacked - Binance got hacked - Polymarket got hacked - GitHub Actions got hacked - SoftBank got hacked - Cursor marketplace got hacked - Coinbase vendor got hacked - Fireworks AI got hacked - LiteLLM got hacked - Capcom got hacked - npm packages got hacked - TiDB Cloud got hacked - Warp's vendor got hacked happy cybersecurity awareness month

strongly disagree with the take that the “AI market is winner take all”. to begin with, if this were true, there wouldn’t be an anthropic. the only reason I could see it being winner takes all is if you plan on “pivotal acts” against all other efforts once you have the capability

Programming is simply lost art now. From the bottom of my heart, I used to love programming, I used to play around with my vim config for hours, learn stupid esoteric languages, then there's stupid fucking nix, all of that -- purely for the love of the game. I cared a lot about the way my code looks, I always wanted it to be pretty, clear, I had strong opinions on how things should be formatted and such. All of that completely unimportant now, it's sad, but I'm glad I'm no longer a part of that.

We reached out to a customer using @render sandboxes to ask about their experience, and they said they had no idea they were using sandboxes. Their agent chose Render sandboxes for their product without asking or informing them, because it was the best tool for the job. This is the future of all product adoption.

The new trend in harnesses is going to be vertical integration btw Harness-defined infrastructure, if you will Your agent harness, orchestration patterns, and infrastructure concerns (like sandboxes) are all collinear Schedules, triggers, event ingestion, sandboxes etc are infra concerns that your harness & orchestration need to have opinions about

There are basically two kinds of companies to build right now. 1. Diffusion of AI This is the "vertical SaaS" of today. There are a bunch of different models for it (FDE company, rollup, horizontal product) but ultimately it all boils down to the same thing: get AI diffused very close to where the work is being done. Profound, Cognition, Harvey etc. A lot of great companies will get built here. 2. Frontier of AI This is where the really interesting action is. Use AI to accelerate scientific progress, then build products from that progress. It is a harder category to see. But we are seeing a lot more of it at @spc: - Externa is adding hundreds of new chemical building blocks to proteins, so you can engineer a drug to match the disease instead of settling for what nature gave you. - Epana Bio is using genomics and AI to design autoimmune therapies that go after the cellular origin of the disease. - VenomicsAI is doing AI-accelerated drug discovery from venoms. - Equivariant is doing multi-physics modeling for the semiconductor industry. (There are always picks-and-shovels for AI as well. Lots of amazing companies in that category.) Category 2 is not well understood or much talked about externally. It should be because I think that is where the vast majority of model driven returns will come from.

We just shipped AWS Emulator, which gives coding agents a local copy of AWS to test against. Early in my career, we built test environments like this by hand, one system at a time. Now, agents are writing code far faster than any team can check it in production. Matt Goldberg, a Senior Principal Engineer on our Developer Tools team, wanted to know whether agents could build and test those environments themselves. He gave @kirodotdev a single prompt, and about 15 minutes later it came back with a first version of an SQS emulator. Turing that into a robust solution took much more work Here’s what I think this means for engineers today: allthingsdistributed.com/202…

In 2004, I picked @SwamiSivasubram as my intern. It was probably one of the best decisions I have made as Amazon's CTO. He learned early that you can only test a system properly in a high-fidelity copy of the environment it will actually run in. AWS Emulator, launched today and written and tested entirely by agents, gives you that kind of environment for more than 200 AWS services. Read Swami's guest post on what the engineer's job becomes when agents write the code: allthingsdistributed.com/202…

Actually, we’re aiming for 1TW/year from Earth, rising to over 1PW/year from the Moon. That’s why I say that if we achieve this outcome, SpaceX will obviously be worth more than Earth, because space is so staggeringly enormous!

I'm the founder and CEO of an 100M+ AI company. Here is the honest truth: 1. AI will make things cheaper. Like the industrial revolution on steroids, most goods and services will go from expensive to essentially free 2. Owners will make lots of money. But most jobs will go away. Which is WHY things will get so cheap Its also why you should go work at a startup that actually gives you ownership