A sentiment shift from prominent engineers who's work defined my generation's childhood is an interesting sign. I myself nowadays vibecode multiple architectures in the span of few days, measure them and pick one which works best.
When I was writing Unreal Engine 1, my time was divided as follows:
- Around 30% went to seriously deep coding on hard problems: assembly language texture mappers, C++ BSP trees, data persistence, object oriented systems design, and networking.
- Around 70% was just typing and refactoring large bodies of code needed for a complete engine feature set. This was not hard, and while I was doing it my mind was mostly on the deeper coding topics, thinking and planning for what's next. It didn't feel much different from vibe coding today, except that it was just me typing code on an IBM Model M keyboard rather than prompting it and reviewing the results.
So, it has always seemed perfectly logical that we would rely on AI for that 70%, once the AI could do it with a similar quality standard as we could, which has never been perfection. It came as no surprise that, after a few years of AI model evolution, it was able to take the reigns on that 70% and free up programmers' time for more valuable tasks.
What's more of a surprise is the emergence of AI as a sort of insta-Carmack able to solve hard technical problems quite cleverly and often cleanly, and that it's able to reason about and work in complex 3D scenes both by operating the tools and by directly understanding the raw data files.
I abandoned assembly language in 1998 when out-of-order CPUs made much simpler C/C++ code faster, and never I missed it. If prompting through architectural documents, concept art, and design plans can transcend typing code and tweaking vertices, that will enable developers to spend their time on higher-level, higher value technical and creative tasks.
What matters is the end product, whether it's a game or an engine. Anything developers can do to try out more ideas and make better software and games faster is a win for everyone.
lmao
I'm so excited to announce a new terminal emulator! 🎉
Meet "Ratty"🐀
🧀 A GPU-rendered terminal emulator with inline 3D graphics.
🪤 Try it out: ratty-term.org
⭐ Source: github.com/orhun/ratty
#rustlang #terminal #ratty #ratatui #opensource
💯
the original TurboQuant paper tested on A100 with models up to 8B.
6 days later, a bunch of strangers on the internet had it built and running on:
- Apple Silicon M1 through M5
- NVIDIA 3080 Ti through DGX Spark Blackwell
- AMD RX 6800 XT and 9070
- a 10-year-old Tesla P40
- an 8GB MacBook Air
- models from 3.8B to 70B across 6 architecture families
- 30+ independent testers
along the way we found new optimizations the paper didn't cover and failure modes it didn't test.
the fact that a loose group of people across the world can read a paper, build implementations from scratch, stress-test across hardware none of us could individually afford, and push the research further in under a week is genuinely one of the best things about this era. the tools and the community make it possible.
open source is something else.
Not all heroes wear capes.
My dear front-end developers (and anyone who’s interested in the future of interfaces):
I have crawled through depths of hell to bring you, for the foreseeable years, one of the more important foundational pieces of UI engineering (if not in implementation then certainly at least in concept):
Fast, accurate and comprehensive userland text measurement algorithm in pure TypeScript, usable for laying out entire web pages without CSS, bypassing DOM measurements and reflow
The thing running on gpu might prefer input to be an image... who would've thought 😅
I quite like the new DeepSeek-OCR paper. It's a good OCR model (maybe a bit worse than dots), and yes data collection etc., but anyway it doesn't matter.
The more interesting part for me (esp as a computer vision at heart who is temporarily masquerading as a natural language person) is whether pixels are better inputs to LLMs than text. Whether text tokens are wasteful and just terrible, at the input.
Maybe it makes more sense that all inputs to LLMs should only ever be images. Even if you happen to have pure text input, maybe you'd prefer to render it and then feed that in:
- more information compression (see paper) => shorter context windows, more efficiency
- significantly more general information stream => not just text, but e.g. bold text, colored text, arbitrary images.
- input can now be processed with bidirectional attention easily and as default, not autoregressive attention - a lot more powerful.
- delete the tokenizer (at the input)!! I already ranted about how much I dislike the tokenizer. Tokenizers are ugly, separate, not end-to-end stage. It "imports" all the ugliness of Unicode, byte encodings, it inherits a lot of historical baggage, security/jailbreak risk (e.g. continuation bytes). It makes two characters that look identical to the eye look as two completely different tokens internally in the network. A smiling emoji looks like a weird token, not an... actual smiling face, pixels and all, and all the transfer learning that brings along. The tokenizer must go.
OCR is just one of many useful vision -> text tasks. And text -> text tasks can be made to be vision ->text tasks. Not vice versa.
So many the User message is images, but the decoder (the Assistant response) remains text. It's a lot less obvious how to output pixels realistically... or if you'd want to.
Now I have to also fight the urge to side quest an image-input-only version of nanochat...
this is cool!
Plasma is better in colour! Watch one of our latest #plasma pulses in our ST40 tokamak, filmed using our new high-speed colour camera at an incredible 16,000 frames per second.
Each pulse lasts around a fifth of a second. What you’re seeing is mostly visible light from the plasma’s edge, glowing pink. The core is simply too hot to emit visible light.
In this footage, lithium is dropped into the plasma. As it interacts, it glows red when excited, then turns green as it becomes ionised, losing an electron. From there, it traces the magnetic field lines, revealing the plasma’s path around the tokamak.
Lithium is the focus of our $52 million ST40 upgrade programme, in partnership with @ENERGY and @energygovuk. This builds on pioneering work at @PPPLab and others that has shown that lithium can significantly improve plasma performance.
This video comes from ongoing research into X-point radiator (XPR) regimes, a promising operating mode for future #fusion power plants that aims to cool the plasma before it reaches plasma-facing components (PFCs), helping to reduce wear without compromising performance.
Fusion research just got a lot more colourful!
👇
tokamakenergy.com/2025/10/15…
#Fusion #FusionEnergy #EnergyTransition #Innovation
I've been Pavel's fan for quire a while now, but after these 4 hours I feel I appreciate him even more. Thanks guys!
Here's my 4+ hour conversation with Pavel Durov (@durov), founder and CEO of Telegram. This was one of the most fascinating and powerful conversations I've ever had in my life.
We discuss everything from his philosophy on freedom to government bureaucracies, intelligence agencies, human nature, mathematics, encryption, great engineering & design, education, family, and his philosophy on life.
It's here on X in full and is up everywhere else (see comment). It is translated and dubbed into Russian, Ukrainian, French, and Hindi.
Timestamps:
0:00 - Introduction
3:07 - Philosophy of freedom
6:15 - No alcohol
14:20 - No phone
20:16 - Discipline
41:28 - Telegram: Lean philosophy, privacy, and geopolitics
56:50 - Arrest in France
1:13:01 - Romanian elections
1:23:56 - Power and corruption
1:33:29 - Intense education
1:45:29 - Nikolai Durov
1:49:58 - Programming and video games
1:54:11 - VK origins & engineering
2:11:24 - Hiring a great team
2:20:40 - Telegram engineering & design
2:39:42 - Encryption
2:44:39 - Open source
2:49:26 - Edward Snowden
2:51:58 - Intelligence agencies
2:53:10 - Iran and Russia government pressure
2:56:19 - Apple
3:03:16 - Poisoning
3:29:28 - Elon Musk
3:35:31 - Money
3:44:23 - TON
3:54:13 - Bitcoin
3:57:12 - Two chairs dilemma
4:03:52 - Children
4:15:02 - Father
4:19:33 - Quantum immortality
4:26:05 - Kafka
