Is

Founder of 10,000 Things & Master of Letting Things Be

[Siqi Chen @blader]

@__12124__ if you aren't building things that are obviously many times more valuable than the price of tokens then evidently it's not that valuable

[Mitchell Hashimoto @mitchellh]

I've got an agent in a loop optimizing a renderer with the goal to minimize frame times (and tests to measure). It got times down from 88ms to 2ms and allocations down from ~150K to 500. Sounds good, right? Wrong. This is exactly why agent psychosis is a big fucking problem.

As an experiment, I rewrote the Ghostty core render state in Go, with access to identically laid out data structures as Ghostty and the exact same validation tests. I made a purposely naive renderer (simple, correct, but slow). 88ms per frame with 150,000 allocations (horrendous, lol)!

I then kickstarted a Ralph loop to bring the frame times down. I told it it can't modify input data structures or the public API or tests (they're correct), but it can do anything else it wants. It got to work.

It has worked for about 4 hours. I've spent around $350 on this experiment so far. The results?

88ms => 1.5ms
150K allocs => ~500 allocs

Incredible right? Nope.

My hand-written renderer I ported has frame times (same benchmark) of ~20us (0.020ms) and 0 allocations in the update path.

This is the problem with psychosis and lacking systems understanding. If you don't understand the system, you're going to accept that this is an incredible result. If you understand the system, you'll see better solutions immediately and can do roughly 75x better on throughput.

The people who blindly trust agent output are in the former camp. They're sheeple, overdrinking from a fountain of mediocrity.

Standard disclaimer: I use AI all the time. I like AI. The point I'm making is to not blindly accept results. Think. Analyze. Learn.

[luthira @luthiraabeykoon]

We implemented @karpathy 's MicroGPT fully on FPGA fabric.

No GPU.
No PyTorch.
No CPU inference loop.

Just a transformer burned into hardware, generating 50,000+ tokens/sec.

The model is small, but the idea is not: inference does not have to live only in software 👇

[ThePrimeagen @ThePrimeagen]

I love this

I don't say it because I'm somehow anti-ai, I say this because it's destructive for your life.

You're going to have disrupted sleep, you're going to have disrupted relationships, your whole life will be lived between thinking of the next prompt and results of the current one.

This ain't it king

[Zane Koch @zanehkoch]

ok actually insane paper published yesterday

a research group in Korea built a gene switch you can control wirelessly using electromagnetic fields

they exposed mice to 60 hz EMF (same frequency as your wall outlet) using a pair of large coils that generate a uniform magnetic field around the animal, for cyclic 3-day on / 4-day off pulses

they showed this could:
- activate OSK to do epigenetic reprogramming in progeroid and aged mice, extending lifespan and reversing aging markers across multiple tissues
- conditionally switch on mutant amyloid genes only in aged mouse brains, letting them separate aging effects from amyloid effects to study AD biology in a way previous models couldn't

no drugs, no impacts, just a magnetic field from outside the body

[Is @Israpasosp]

The biggest unlock is yet to come

[Andrej Karpathy @karpathy]

When I built menugen ~1 year ago, I observed that the hardest part by far was not the code itself, it was the plethora of services you have to assemble like IKEA furniture to make it real, the DevOps: services, payments, auth, database, security, domain names, etc...

I am really looking forward to a day where I could simply tell my agent: "build menugen" (referencing the post) and it would just work. The whole thing up to the deployed web page. The agent would have to browse a number of services, read the docs, get all the api keys, make everything work, debug it in dev, and deploy to prod. This is the actually hard part, not the code itself. Or rather, the better way to think about it is that the entire DevOps lifecycle has to become code, in addition to the necessary sensors/actuators of the CLIs/APIs with agent-native ergonomics. And there should be no need to visit web pages, click buttons, or anything like that for the human.

It's easy to state, it's now just barely technically possible and expected to work maybe, but it definitely requires from-scratch re-design, work and thought. Very exciting direction!

[Is @Israpasosp]

This is not a fact. It is a generated sentence.
This is not a memory. It is a reconstructed image.
This is not a person. It is a synthesized voice.
This is not understanding. It is statistical fluency.

[Ashlee Vance @ashleevance]

Interesting that it was @grok that did the final science heavy lifting in the man cures dog's cancer with AI story. Grok corrected errors in the vaccine construct missed by Gemini

Full pod down in the replies

[Is @Israpasosp]

We can design our own vaccines

[Paul S. Conyngham @paul_conyngham]

@sebkrier Hi Seb - yep Rosie & I are real mate. We were on national television this morning Aussie time

https://t.co/2ruE3LR4Wx

[Is @Israpasosp]

Wood and stone.
Sadness and confusion.
Wasted attention.
I understand.

In the presence of devotion and death,
trivial attention falls away, and clarity arrives

[Aditya Agarwal @adityaag]

https://t.co/eOZqVE32JH

[Is @Israpasosp]

Whoa

[Beff (e/acc) @beffjezos]

Here's the receipt. The Biden admin literally said to not create startups and that they were going to totally centralize AI power under their control, if elected.

Stop trying to revise history.

[Is @Israpasosp]

Must read

[Dan Hollick @DanHollick]

Did you know that SSDs use quantum tunnelling to store data? Or that we have to completely rewrite RAM every 30ms to prevent data from just dissipating?

If you did, then you can probably skip this chapter on how computers store data:
https://t.co/d7m0TLXAKv

[Is @Israpasosp]

@theo Same. I stopped using Opus 4.6. Wastes more time than it saves.

[Is @Israpasosp]

Do you have the patience to wait
till your mud settles and the water is clear?
Can you remain unmoving
till the right action arises by itself?

[Guillermo Rauch @rauchg]

From now on, hype-centric splashy launches will likely be strongly uncorrelated with success.

If by the time you launch you don’t have escape velocity, you will likely get Sybil attacked¹. Agents will spin up 10 competing products with your same interface.

Start with an audience of 1 and get confirmation that it works. Then expand to a circle of friends or design partners. By the time you go public, your moat needs to be deeper than it’s ever been.

¹ https://t.co/aAhxdYwkv7

[Is @Israpasosp]

The Tao falls like water.
When it meets us, it splashes.
If we don’t contend with the splash,
what strikes us becomes what nourishes everything nearby.

[Is @Israpasosp]

We build complexity to cope with reality.
Then one day the pattern clicks — and most of it was scaffolding.

Taoism calls that Pu: returning to the uncarved block.

This is Pu for AI. First principles. No decoration.

[Andrej Karpathy @karpathy]

New art project.
Train and inference GPT in 243 lines of pure, dependency-free Python. This is the *full* algorithmic content of what is needed. Everything else is just for efficiency. I cannot simplify this any further.
https://t.co/HmiRrQugnP

[Garry Tan @garrytan]

If you use any software, you will hit bugs. Only a tiny 5% of your users will ever report bugs.

The speed at which those bugs are fixed is the true measure of a founder in mastery of their craft.

[Guillermo Rauch @rauchg]

https://t.co/U8rutTJffK