AI Reading List: April 2026

A few links from recent weeks that I found interesting:

Highlights from my conversation about agentic engineering on Lenny’s Podcast - An AI “State of the Union” with Simon Willison. Agree with most everything here. If you’re just going to read/listen to one thing from this list, this is the one.
The Shape of the Thing — Another nice State of AI post From mid-March.
Eight years of wanting, three months of building with AI — Retrospective of Pros/Cons of Agentic AI for a large project. Key insight: AI is a force multiplier for implementation but a dangerous substitute for design.
Claude Cowork Gave Me AI-Induced Mania — Clickbait-y title, but the discussion of the scaffold (markdown files for context, preferences, workflows) makes this worth the read.
The Cult of Vibe Coding Is Insane — Fairly critical analysis of the Claude Code leak from a creator of BitTorrent. My take is that if Claude Code is this useful as an messy, early beta imagine how good it’s going to be in a year or 2 …
The Beginning of Programming as We’ll Know It — A measured take on the confirmation bias in AI success stories and the programmer’s role in agentic workflows.
Components of a Coding Agent — Nice detailed semi-technical explanation of how coding agents work.
Caveman: why use many token when few do trick — Cute plugin that makes your Agent talk like a caveman (thus saving tokens).
Kyle Daigle on GitHub’s surging activity — A few eye-popping stats regarding GitHub Actions usage from GitHub’s COO.

Agentic Pattern: First Run the Tests

The Capital Might Be Misaligned, But The Plane is Probably Fine

Comments