kottke.org posts about artificial intelligence (kottke.org)
Posts about artificial intelligence on kottke.org — links and commentary by Jason Kottke.
As an Experienced LLM User, I Actually Don't Use Generative LLMs Often
minimaxir.comBut for what I do use LLMs for, it’s invaluable.
Vibe prototyping isn't solving problems. It's creating new ones.
It's easy to prototype and so everyone is prototyping, without really knowing why they are doing it. The resultant noise drowns out thoughtful work.
How I don't use LLMs
Manfred Mohr, Cubic Limit: P-197 (1977)I enjoy shocking people by telling them I don’t use LLMs.This isn’t true, but it’s morally true for the reference cl...
API vs. chatbots? They're singing from the same hymn sheet
They have their pros and cons but they're not directly opposed. The key difference is manual fine-tuning.
Much Ado About Vibe Coding
Lauren Goode convinced her editors at Wired to let her spend a couple of days at a tech company called Notion learning how to vibe-code (i.e. AI-assisted comput
LLM Friendly Projects | David Gasquez
David Gasquez personal website
Good Old Fashioned Linkblog
The personal site of R. W. Blickhan
The Skeptic Who Got Benchmaxxed: What Actually Changed About AI Coding Agents | Rezha Julio
I used to dismiss AI coding agents as expensive autocomplete. Then Opus 4.5 dropped, and a data scientist started shipping Rust crates that beat numpy. A breakdown of Max Woolf's journey and why the AGENTS.md file might be the most underrated tool in the stack.
Hacker Bits, Issue 116 - Hacker Bits
Welcome to issue 116 of Hacker Bits! We have a great selection of articles this month covering browser fingerprinting, SQL Noir, writing design docs, DeepSeek-R1, vibe coding, AI hate, tricks of the trade, and so much more! As always, please ... Read More
uses for LLMs
In this post: specific use cases for LLMs that seem to work and are reasonably useful. "LLM to improve my writing without having it do my wr...
The universal translator
A few months ago, I wrote about the deep weirdness of LLM technology, and how we’re all still learning how to approach its outputs. The marketing and popular perception of these models is that of a helpful robot assistant, and leaning into that characterisation is also a core part of what makes the technology useful in ways that straightforward text completion models weren’t. By pretending to be a robot assistant, modern instruct models/chatbots are able to perform many digital tasks one might want a (digital) robot assistant to perform.
Organizing a design system via folksonomy
No hierarchy other than what people create for themselves.