On architecture decisions, AI agents, composable commerce, and the gap between strategy decks and what actually ships.
The Model Context Protocol is elegant on paper. In practice, connecting AI agents to enterprise systems surfaces a set of problems the specification was never designed to address.
The Model Context Protocol is elegant on paper. In practice, connecting AI agents to enterprise systems surfaces problems the spec wasn’t designed to address.
Most re-platforms don’t fail because of technology. They stall because nobody aligned on sequencing, domain boundaries, or what “done” looks like for the business.
By the time someone calls for an architecture review, the damage is usually done. Here’s the case for building review into your operating rhythm — not your escalation path.
Teams adopt event-driven architecture and then try to make everything an event. That’s how you end up with a distributed monolith that’s harder to debug than the one you replaced.
The hardest part of architecture at scale isn’t the system design — it’s getting 40 people to share a mental model of why the system works the way it does.
No schedule, no filler. If you want to know when something new goes up, the best way is to follow me on LinkedIn.