I love prototyping. It’s the most important thing that sets great software above the merely good. And to that end, here’s a project a couple of us have been hacking away on, and the team is ready to add people in mass to the TestFlight. It’s an open source version of a visual programming environment, similar to Origami or Quartz Composer. In addition to being able to build anything those two can, it also supports AR, CoreML, and native 3D support. As well as being built in SwiftUI, so it runs on basically any Apple device and is easy to hack on and improve.

Introducing Stream & Stream Ring. Let thoughts & ideas flow ∽ Preorder now at sandbar.com

An under-discussed topic: how the hottest software engineering job of the early 2010s is seeing a steady but ongoing decline the last few years. I'm talking about the native iOS and Android positions. Outside of Big Tech, few startups/scaleups hire for this. Since ~2022?

The reason everyone in sf works so hard is they have all “fumbled” great opportunities to work on what matters to them

Made a macOS app at work this week just to try something out. Made it purely with Claude Code and a handful of manual tweaks. There’s no excuse to not be prototyping in native material these days.

📣Announcing the first preview releases of Swift for Android, enabling you to build Android business logic with the same Swift that you use for Apple platforms. swift.org/blog/nightly-swift… #Android

What happens when you turn a designer into an interpretability researcher? They spend hours staring at feature activations in SVG code to see if LLMs actually understand SVGs. It turns out – yes~ We found that semantic concepts transfer across text, ASCII, and SVG:

Palmer Luckey (@PalmerLuckey) is getting into nicotine. "Basically, America smoked its way to being the dominant hyperpower. It kept people focused, it kept people fit. It's an appetite suppressant. "I'm becoming more and more convinced that the health benefits of not smoking have not been properly traded against the health problems caused by the resulting eating." "I think the crazy version of this is we'd all be better off with lung cancer eventually, but fit until then." "Is vaping worse than being fat? I don't know. I struggle. Based on the evidence I've seen."

What SF Compute does. When you finance a GPU cluster, you need to get an "offtake" agreement. Basically, someone has to agree to rent the cluster from you, typically for a 3+ year period. If that agreement falls through (the person fails to pay), then the person who owns the cluster gets wiped out, and their lender ends up with a bunch of GPUs, rather than say, money. It really looks like the world is deploying more capital into the AI build out than any infrastructure project in the history of the world. You remember when people said there was going to be a Manhattan project for AI? The current build out is the size of 20 Manhattan projects. We’re so far past the Manhattan project it’s not even funny. This is the cost of a war. It would be really bad if that scale of capital was secured against offtake agreements (long term contracts) with application layer companies who turn around and sell to their customers on a month to month basis. If the AI SaaS has a bad few months, can the AI SaaS continue to front their compute bill? They could in CPUs, because in SaaS you might have a company with $20m in the bank, and has a $1m/year "CPU" bill. But in GPUs, you have a startup that raised $20m, but a $20m+/year compute bill. So a small shift in demand means lights out for your business, because the products are so levered. That works as long as you can plug the gaps with venture capital & high margins. But across the board, AI applications are lower margin than their SaaS counterparts, giving them less buffer to save them in a bad month. And even in a hot market, venture capital won't necessarily save you if you're running unprofitably with a massive liability. That’s the problem we solve. We let people buy long term contracts they can “exit”, by selling back. That lets them get liquidity in the most critical moments, ensuring they turn a profit rather than a loss on tight margins. In other words, we prevent a bubble. When we do that, it opens up blocks of compute for smaller use cases too, like academics or startups. When we started, we were "Junelark", a 2-person audio model company that bought too big of a cluster. We had bought 12 months, but could only afford 1 month. To avoid bankruptcy, we had to sublease the other 11 months by acting as GPU brokers. Our audio model company was forced to pivot or die because we didn't have liquidity. To make SF Compute, we split the company down the middle. One side of the house makes a fintech company, a ledger, an order book, and a compliance program. The other side makes a systems engineering company. To make this work, you need to run the clusters. So we make the low level cloud stack that interacts with BMC (Redfish & IPMI), UFM, built a UEFI app that replicates PXE boot in weird environments, and a virtualization layer kind of like EC2. It’s a massively complex machine filled with nitty gritty challenges. Today, we’re growing faster than Cursor and we’re scaling to secure the risk of the largest infrastructure build out in the history of the world. We’re hiring across the board for rust programmers, systems engineers, and GTM, and we’d love for you to join us to prevent an AI bubble.

Today @GoogleDeepMind released a state of the art computer use model, in partnership with Browserbase. Computer use is hard to evaluate. You need reliable browser infrastructure and realistic tasks. Below, we cover how we ran these benchmarks and how you can try yourself!

New paper 📜: Tiny Recursion Model (TRM) is a recursive reasoning approach with a tiny 7M parameters neural network that obtains 45% on ARC-AGI-1 and 8% on ARC-AGI-2, beating most LLMs. Blog: alexiajm.github.io/2025/09/2… Code: github.com/SamsungSAILMontre… Paper: arxiv.org/abs/2510.04871