Daily Reading List – November 18, 2026 (#912)

Jetlag finally hit me today. I’ve been dragging all day, which isn’t ideal midweek when a ton is going on. At least it’s already dark outside. That’ll perk me up.

[blog] Introducing Gemini 4 Neon. Can’t stop, won’t stop. We’re delivering frontier model after model, with this one fitting that middle sweet spot.

[article] The Codebase Rewrite is Back. Are we going to break one of the unwritten rules in tech and embrace rewrites versus update-in-place? The economics and level-of-effort changed, so yes.

[blog] Introducing Stitch Loop: A new way to build and evolve products. Super interesting. Maybe your dark (or light) factory needs this trigger that offers continuous improvement suggestions for your web application.

[blog] How to Speed Up CI in the Age of AI-Written Code. Good list. If you’ve procrastinated on fixing slow-ish CI pipelines, now’s the time to invest.

[blog] Transcript Testing: Helping Humans Verify AI Code. Shouldn’t humans still be able to read the code and tests? For now, that’s important. Here’s the thought of creating a transcript of the system’s behavior that turns into a test.

[blog] TPU Model Design Considerations and Optimization Guide. TPUs are popular, and increasingly so. This post has key principles to accelerate workloads on TPUs.

[article] The 10x org is in your grasp. Maybe! At least for categories of engineering work. I’m producing 3x at least myself, but that’s just for output. My meetings and face to face work can’t 10x.

[blog] Announcing A2UI 1.0 and a2ui_flutter. A2UI supports many renderers, and one of them is Flutter. Check out this post to get a real sense for when to use generative UIs, and how to get started.

[article] Microsoft’s new coding model promises near-Opus performance at 5% of the cost. Cost matters when using these models at scale. Maybe even more than performance.

[article] How do AI code review agents spend their effort? How do measure the success of a code review agent? What role does cost play? Some direction here.

[blog] Goal Engineering: How to Design Goals. Goal-setting is important if you’re unleashing AI on some work task. This post encourages us to aligns goals to what’s easiest for the human to review.

[blog] Running Untrusted AI Agents on GKE Agent Substrate: Sub-Second, Zero-Idle, and Sandboxed. Wonderful walkthrough of why Agent Substrate matters, and then exactly how to use it.

[blog] How to Deliver Projects With Unclear Scope. AI doesn’t fix this. It’s just hard work. This post has a story and some advice.

Want to get this update sent to you every day? Subscribe to my RSS feed or subscribe via email below:

Comments

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.