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
Caught up with @karpathy for a new @NoPriorsPod: on the phase shift in engineering, AI psychosis, claws, AutoResearch, the opportunity for a SETI-at-Home like movement in AI, the model landscape, and second order effects
02:55 - What Capability Limits Remain?
06:15 - What Mastery of Coding Agents Looks Like
11:16 - Second Order Effects of Coding Agents
15:51 - Why AutoResearch
22:45 - Relevant Skills in the AI Era
28:25 - Model Speciation
32:30 - Collaboration Surfaces for Humans and AI
37:28 - Analysis of Jobs Market Data
48:25 - Open vs. Closed Source Models
53:51 - Autonomous Robotics and Atoms
1:00:59 - MicroGPT and Agentic Education
1:05:40 - End Thoughts
📽️ From Donald Trump to Britain's wind power trade body, there's a growing coalition calling for more drilling in the North Sea.
Raising the question: if we DID encourage more exploration, how much oil & gas could we actually get?
Our MEGA primer on the North Sea👇
Ps it's longer than usual, but it turns out this topic has SO MANY misconceptions. Time to put some of them right.
Let me know what you think
My take on current state of LLM coding models.
@cline with sonnet 4.5 is unbelievable.
Codex high is thoughtful and very good at BE work, but surprisingly poor at FE.
My general flow is cline + Claude to plan, then ask codex to implement backend and cline + Claude for FE