What makes software easier for AI coding agents to understand, maintain, and change? In episode 139 of Betatalks, Rick and Oscar welcome Hannes Lowette to explore event sourcing, domain-driven design, and the architectural choices that help both developers and AI agents work effectively.

Hannes explains how recording meaningful business events—and separating command handlers, projections, and downstream actions—can keep the context of a change small. He shares why this structure could make event-sourced systems a strong fit for AI-assisted software development, while discussing when event sourcing adds value and when it introduces unnecessary complexity.

The conversation covers microservices, modular monoliths, and the pitfalls of choosing architecture based on industry trends rather than business needs and team experience. They also explore AI-assisted legacy code migration, why code quality still matters, and how clear naming, specifications, tests, and Architecture Decision Records (ADRs) support better results.

Beyond the technical discussion, Hannes shares his approach to coaching, knowledge sharing, and conference speaking. Plus, discover how woodworking, a handmade travel guitar with a Visual Studio inlay, and performing with The Linebreakers connect him to the developer community.

An episode for software developers, architects, and technical leads thinking about maintainable systems and the role of AI agents in software engineering.

Follow Betatalks for more conversations about software development, .NET, AI, and software architecture.

About this episode, and Hannes Lowette in particular: you can find Hannes on LinkedIn.

About Betatalks: watch our videos and follow us on Instagram, LinkedIn, and Bluesky.