🧱 Architecting stuff so it doesn’t fall apart
🧪 Frontend, platforms, AI-ish things
🎿 Skiing when things work, debugging when they don’t
The Shift Has Begun: Why Software Architecture Needs to Change for AI
AI isn’t just a feature. It’s changing how software is built, how it behaves, and how we reason about the systems behind it.
If you’re still designing services like it’s 2018, you might be building a house on sand. AI workloads are modular, metadata-driven, and dynamic in ways most traditional architectures don’t account for.
In this post, I break down what’s changing and why—and explore a few ideas for how to rethink architecture in the age of agents, fine-tuning, and system feedback loops.
Read the full piece here:
👉 https://medium.com/software-architecture-in-the-age-of-ai/the-shift-has-begun-why-software-architecture-needs-to-change-for-ai-d84e73a1cdf8
🧠 Key themes:
- Architecture as metadata
- Systems designed for coordination with AI agents
- The rise of modularity and introspection as first-class needs
- Real-world examples from the trenches
Would love to hear how your teams are adapting. Are you already rethinking services for AI? Or still patching workflows that were never meant to be intelligent?
Drop a comment. Let’s build the future like we mean it.
📚 Part of the series:
Designing for Intelligence – Software Architecture in the Age of AI