Claude Opus 4.6 and agent swarms were announced yesterday. That's some big upgrades for Claude Code.

OpenAI, the competition, offered us GPT-5.3-Codex, and this week gave us an app form of Codex that already has a million active users.

That's all very exciting, and next week is going to be about covering that.

This post is about all the cool things that happened before that, which we will be building upon now that capabilities have further advanced. This if from Before Times.

Almost all of it still applies. I haven’t had much chance yet to work with Opus 4.6, but as far as I can tell you should mostly keep on doing what you were doing before that switch, only everything will work better. Maybe get a bit more ambitious. Agent swarms might be more of a technique shifter, but we need to give that some time.

Table of Contents

---

Outline:

(01:04) Claude Code and Cowork Offer Mundane Utility

(04:21) The Efficient Market Hypothesis Is False

(07:51) Inflection Point

(11:40) Welcome To The Takeoff

(12:04) Huh, Upgrades

(16:51) Todos Become Tasks

(18:36) I'm Putting Together A Team

(21:00) Compact Problems

(21:49) Code Yourself A Date

(25:20) Verification and Generation Are Distinct Skills

(27:11) Skilling Up

(36:44) AskUserQuestion

(37:14) For Advanced Players

(39:33) So They Quit Reading

(40:07) Reciprocity Is The Key To Every Relationship

(44:19) The Implementation Gap

(47:57) The Lighter Side

---

First published:

Source:

---

Narrated by TYPE III AUDIO.

---

Images from the article:

Apple Podcasts and Spotify do not show images in the episode description. Try Pocket Casts, or another podcast app.