AI, data and resilient infrastructure for environments where operations matter
I work at the intersection of infrastructure, data, AI and operational resilience. My focus is environments where latency, uptime, security and continuity have real consequences, and where good decisions depend on a foundation that can be trusted.
read the writing ›AI does not remove the need for architecture. It increases the need for it.
In operationally critical environments, AI depends on reliable data, strong infrastructure, observability, process maturity and clear ownership. The model is rarely the hard part. The foundation is.
AI writes the code now, and it writes code that compiles, passes the tests, and reads as though a competent engineer wrote it. None of those is a security property. The writing got cheap. The reading did not, and the reading is the control.
read the essay ›The price is not the cost
Failover assumes a copy
Fluent is not the same as correct
The model inherits the foundation
Build observability for the worst moment
The interesting problems are rarely about a single piece of technology. They are about whether the foundation underneath is good enough for the thing you are trying to build on top of it. A few things I believe after having to keep these systems running, not only design them:
- Reactive IT is not cheaper. It moves cost into incidents, rework and risk.
- Observability is not dashboards. It is decision support.
- DNS and private endpoints are cloud foundations, not minor technical details.
- Problem management is how an organisation stops treating every incident as new.
- Good AI work needs infrastructure, ownership and operational reality, not just a model.