Status update: I've been on/off AI agents in the last few days and it is a verifiable truth that every day I didn't use agents, I was more productive. I still attribute that to how slow they are, and my own inability to multi-task efficiently. The magic is there but the slowness doesn't let it cross the threshold where they actually make me faster, and I still dislike the whole thinking paradigm. About Bend2: honestly, the C/Metal compiler codebase is a clusterfuck right now. I regret letting AI agents write it. All tests pass, and GPU performance is mind-blowing, so the core architecture works. Yet, it has a LOT of bugs. Anything not covered by the tests is a coin toss. This is actually impressive, because, in many parts of the codebase, the right solution was actually the simplest one, yet, the agents STILL managed to find a way to make it work just for the tests. The level of reward hack these agents output is actually impressive I can't even be mad. It is also ironical because that's the very problem that Bend's proof system was supposed to solve, but Bend is in TypeScript, not in Bend. I'm disappointed I didn't write Bend in itself, and now I feel an immense urge to do so. But the clock is ticking . . . Still, I do not think Bend is worth launching without the GPU compiler being solid, because the closest competitor, Lean, is actually extremely good, so we need a big differential. Yet, due to the very nature of the project, it would be embarrassing to have bugs at launch. Regarding AI, I now believe using current gen AI agents in production codebase is harmful and a massive mistake. That doesn't mean no agents at all, but agents work best when they don't touch critical code. Debugging, researching, providing insights, scripts / tools, or anything that doesn't touch code you will maintain in the long term. But if you merge AI code without reading, you're going to have a bad time. Speaking from experience I'm working 10h/day on SupGen and the remaining time on Bend2

The AI panic is really unbelievable today. The level of delusion and hype have grown to mythic proportions. Has AI beaten Pokemon Red yet? Like a normal 6 year old does, by looking at the screen? Oh it hasn't. But all jobs are over in 18 months? This website is full of idiots.

its not even been a day and i am already addicted. the cat is the cherry on top

uv has 350k lines of Rust, and 73 "unsafe" calls. The Bun Rust port is already 681k lines of Rust, and has over 13,000 "unsafe" calls.

OVERRATED: running tons of agents in parallel; working on too many things at once; perpetual context-switching; opening lots of low-quality PRs that may never land. UNDERRATED: using one or two agents at a time; focusing on the task in front of you; thinking deeply; finishing stuff; making your code works in prod.

opus 4.7 is a beauty a fresh yet elegant take on something we've seen before a new standard, a definite marker of a new era (i haven't tried it yet)

this is really bad i hope this gets everyone to stop and reflect on how poorly this is all going ai is incredible tech that every single person has had at least one incredible experience with for it to be so hated despite that is a huge failure

slopped up a basic skill for @badlogicgames Pi and @DanielGri's Glimpse to organize my diffs into sections for a Devin Review-like experience. will probably polish it up and publish soon :)

as one of those kids who was building with and contributing to Pi before that was cool, this seems like the right move. best of luck to mario and the earendil team :)

the general purpose harness is actually the one that's shitty out of the box and let's you turn it into a lean mean task machine by just telling it "do the thing, make no mistakes"

conspiracy by openai to artifically degrade gpt performance after 272k tokens to save VRAM

🚨BREAKING: Jared Issacman has allegedly asked Lockheed Martin to equip the Artemis IV Orion capsule with a space laser to “destroy all opps”

While you are building AGI, top article on HN is how to make all the corner radii the same on macOS

there was a moment in 2023 when your team was using Figma, Linear, VS Code, Typescript, React 18, esbuild or Next.js with the pages router. Peak software in every category. wonderful stack. you were young and happy and in love

A glimpse of the absolute insanity in numbers that is our industry right now. Attributed Claude activity over the last 90 days on GitHub. The pushers did us dirty over Christmas.

i have a nagging feeling in the back of my head that, as models get more powerful, autoresearch-style tasks will slowly just become prompts rather than needing specialized harnesses. same thing happened to ralph loops imo

I'm usually not one to write thought pieces without much technical depth. But here we go. Slow the fuck down. mariozechner.at/posts/2026-0…

who's speedrunning the pi extension for this? if not I might just have to do it myself...