Category Archives: Software Architecture

Part 2: Sagas, Fan-In, and Correlation: Solving the Hard Problems of Eventual Consistency in EDA

Choosing event-driven architecture means choosing eventual consistency. The only question is how well you manage it. When payment succeeds but inventory reservation fails, do you refund automatically or let a customer stare at a half-completed order? This guide covers Saga compensation patterns, the callback topic debate (and why it’s really async RPC over event infrastructure), fan-out vs fan-in complexity, and three concrete approaches to multi-topic correlation — with production code in Python, Java, and AWS.

Before You Adopt Event-Driven Architecture: Prerequisites, Red Flags, and Partition Strategy

Most teams adopting event-driven architecture never stop to understand what it actually involves. Before committing to the operational overhead that transforms a three-component system into fifteen, verify you actually have the problems EDA exists to solve. This guide provides a decision framework, prerequisites checklist, and a deep dive into partition strategy — the most consequential design decision you’ll make in EDA.

AI Is Forcing Architects To Redefine How We Validate Software

AI exposes bottlenecks in manual review, vague requirements and outdated validation models.
This article explains how spec-driven development, architectural truth models and agentic CI/CD pipelines reshape delivery for the AI era.

Part 3 – Inside an AI-First Pod: How Four People Out-Deliver a Team of Twelve

You don’t believe it until you watch it happen. Four people. One requirement. Four hours later, it’s in production. The first time you see it, you assume they cut corners. Test coverage? 94%. Error handling? Comprehensive. Security? Passed automated checks. They didn’t cut corners. They just removed the waiting. See exactly how a pod operates—including what happens when the architect disappears for three weeks.

Part 2 – The AI-First Delivery Model: Why Your Agile Team Can’t Just Add AI and Hope

AI is no longer a convenience tool. It has changed the pace of what is possible, while most organisations kept the same delivery model. This article explains why effort no longer wins, why structure does, and how AI first pods can ship in hours instead of weeks.