#ai

21 posts

Your traces are the best context window an agent will ever get

Keeping your traces close to your e2e testing execution, and your feedback cycle as short as possible is more important than ever.

Oguzhan Soykan

Mar 12, 2026

Compile to Architecture

For a long time we’ve treated frameworks as the target of software development. But if systems are meant to be regenerated and replaced safely, the real compilation target has to be the architecture itself.

The Phoenix Architecture

Mar 6, 2026

The Industrialization of Regenerative Software

The “AI software factory” metaphor is seductive.

The Phoenix Architecture

Feb 11, 2026

Douglas Adams on AI, Before There Was AI

A passage from one of the author’s comic novels of the 1980s clarifies a key user flaw of our AI present

Journey to the End of the Shelf

Jan 31, 2026

On AI Contribution Policies

You should have one, and you should say "no".

Alexia's Ocassional Ramblings

Jan 30, 2026

Muckin’ Around: An AI Manifesto

Why start a blog in 2026 anyway...

Muckin' Around

Jan 28, 2026

The Broken Trust Bridge: How AI Shattered the 70-Year CV and How to Rebuild Hiring

The foundation of the hiring process has been irreparably changed. For 70 years, the Curriculum Vitae (CV) served as the primary trust mechanism between candidates and employers. It was a tangible work sample that demonstrated writing ability, organizational skill, and attention to detail, creating a starting point for an honest dialogue.

The Broken Trust Bridge: How AI Shattered the 70-Year CV and How to Rebuild Hiring

Jan 26, 2026

The Deletion Test

Here’s a simple test you can apply to any software system you work on:

The Phoenix Architecture

Jan 23, 2026

UI Is a Conservation Layer

Why the user interface is the last to become regenerative

The Phoenix Architecture

Jan 21, 2026

Provenance Is the New Version Control

When code can be thrown away and recreated, the unit of change is no longer lines of code. It’s reasons. Version control has to follow.

The Phoenix Architecture

Jan 12, 2026

n=1 Is a Design Constrain (Not a Staffing Model)

Single-developer capability isn’t a productivity story. It’s the test that tells you whether your architecture is worth keeping.

The Phoenix Architecture

Jan 7, 2026

Relocating Rigor

The Discipline That Looks Like Recklessness

The Phoenix Architecture

Jan 6, 2026

The System Is the Asset

Why Regeneration Does Not Mean Starting Over

The Phoenix Architecture

Jan 5, 2026

Conceptual Mass and the Compaction Discipline

As I mentioned in a previous post, at Wunderlist, we had a rule: any new service had to be "this big", a constraint I'd demonstrate by holding my fingers a few inches apart. The metric wasn't about lines of code. It was about replaceability.

The Phoenix Architecture

Jan 2, 2026

Immutable Infrastructure, Immutable Code

Why "Never Upgrade in Place" Now Applies to Software

The Phoenix Architecture

Dec 30, 2025

Evaluations Are the Real Codebase

Why behavior outlives implementations

The Phoenix Architecture

Dec 29, 2025

The Gradient of Trust

Better shapes beat better prompts

The Phoenix Architecture

Dec 28, 2025

Code Was Never the Asset

Why AI makes the hidden economics of software unavoidable

The Phoenix Architecture

Dec 24, 2025

Pace Layers and AI Integration

Not all software should change at the same speed.

The Phoenix Architecture

Dec 23, 2025

The Death and Rebirth of Programming

Programming didn't die all at once. There was no single moment, no dramatic obsolescence event. Instead, something quieter happened: the core constraint that shaped software for seventy years dissolved. Writing code stopped being the hard part.

The Phoenix Architecture

Dec 22, 2025

Regenerative Software

Software is entering a strange new phase.

The Phoenix Architecture

Dec 21, 2025