How can an LLM write code?

It can feel like AI is thinking. What’s actually happening is closer to pattern recognition - and that frame makes LLMs much easier to reason about.

March 29, 2026 · 1 min · 197 words · Agoston Fung
Illustration of a person typing on a laptop in polite chat with a friendly AI - speech bubbles read 'Please assist me with that', 'Thank you!', 'Of course!', 'Here's the information', and the robot replies 'Happy to help!'. To the side, an angry-face emoji and a haloed-face emoji represent the choice of disposition. Underneath, a small brain-to-robot transformation visual.

Autocomplete for your personality

Sam Altman said millions are wasted on people thanking LLMs. I keep doing it anyway. Here’s why - and what your brain does when you don’t.

March 10, 2026 · 1 min · 175 words · Agoston Fung