AI Engineering Transformation
What Is Your AI Engineering System Optimising For?
Adoption tells a board how much AI it bought. Before you redesign engineering around agents, agree the one number the redesign has to move.
Tag
Production Systems
Articles exploring production systems.
AI Engineering Transformation
AI Won't Replace Your Engineering Bottleneck. It Will Move It.
Every stage you speed up with AI hands the constraint to the next one. The useful question for a CTO used to be where the bottleneck is. Now it is where it goes next, and how fast.
AI Engineering Transformation
Context Engineering Is the Wrong Altitude
The industry has renamed the practitioner's job twice—from prompt to context to harness engineering. Each rename is accurate, and each stops at the edge of the agent. The constraint has moved into who owns, funds and governs context across the organisation.
AI Engineering Transformation
The Second Absorption Gap: When Agents Outrun the Context You Can Govern
AI is putting agents into production faster than organisations can supply and govern the context those agents run on.
AI Engineering Transformation
The Absorption Gap: Why Your AI Adoption Metrics Are Lying to Your Board
AI is increasing engineering supply faster than organisations can absorb it. The real job of the CTO is shifting from making engineers faster to redesigning the system that now has to keep up with them.
Systemic CTO
Do Not Fix Tech Debt. Fix the System That Creates It
Architecture debt, incentives, team boundaries and skills concentration.
Quality Engineering
7 Signs You Need a Systemic Approach to Software Production
The indicators that the production system, not the teams, is the constraint.
Enterprise Architecture
Accelerate Monolith Deployments Without Microservices
From three months to three days on a business-critical legacy system.
Quality Engineering
MAMOS: Architecting the Software Production System for Quality at Speed
A capability map of methods, architecture, management, organization and skills.
Quality Engineering
Quality Engineered, Speed Delivered
Why quality practices alone slow you down, and what changes when you treat production as a system.
Quality Engineering
Zalando’s Quality Engineering Journey: From Monolith to Microservices
How Zalando rebuilt its architecture, management, organisation and skills to sustain growth, with positive impacts on innovation, productivity and growth.
Enterprise Architecture
How Google Uses a Monorepo: Tooling, Testing and Trade-offs
How Piper, trunk-based development, code ownership and automated testing make Google’s monorepo workable—and what that model costs.
AI Engineering Transformation
The AI Engineering Production System: Designing an Organisation for Agentic Software Development
Everyone now agrees AI means redesigning engineering. Few say what to redesign, in what order, or how to tell it worked. This is the map: a frame, six stages, a workforce of agents and people, and one layer under all of it.
AI Engineering Transformation
The Agent Ready Delivery Pipeline: What Survives When Change Volume Explodes
Once verification keeps pace with agents, the constraint moves to integration: the main branch, the environments and the test data. This is the pipeline design that held at La Redoute, and why it is what agent volume now needs.
AI Engineering Transformation
Designing the Verification System: The Architecture That Lets You Safely Absorb Agent Output
Review cannot keep pace with agents, and the data suggests organisations are quietly lowering the bar instead. This is what to build: requirements you can test, checks where work is produced, risk tiers, and a release decision kept apart from generation.
AI Engineering Transformation
Building a Governed Context Supply: What Your Agents Need to Know About Your Systems
Agents act on whatever your organisation has written down about its systems. Most of it is stale, unowned or scattered, and adding more of it does not seem to help. This piece sets out a context supply that does, and how to measure it.