Cloud economics
Turning complex AWS, Azure, and GCP infrastructure spend into decisions and daily cost accountability.
About
The economics of compute, from the cloud bill to the chip.
I'm a cloud economist and ML engineer examining what it costs, in dollars and in watts, to deliver a unit of intelligence.
The work
A defensible view of cost, risk, and outcome.
I work with FinOps, platform, SRE, and engineering leaders — and the decision makers they support — to make AI cost and risk legible. A vendor rate or polished dashboard is not enough. They need a defensible view of what a system costs, what it exposes, and whether the outcome justifies the spend.
I bring FinOps discipline across AWS, Azure, and GCP to AI systems. I trace the full stack — from inference and agent controls to hardware, energy, and thermal effects. Cost per token is only the starting point. The more useful question is the cost, in dollars, watts, and risk, of an outcome the organization can defend.
I work as an independent second opinion. I build the models and tools behind the analysis, including inference-cost and remote-sensing systems, so technical teams and decision makers can act on measured evidence.
Areas of practice
Following the full system.
AI system economics
Tracing inference, agent controls, hardware, energy, and risk to understand the cost of a defensible outcome.
Independent analysis
Building measured numbers, models, and tests that operators can take into consequential decisions.
Applied ML systems
Building the inference-cost models and remote-sensing systems behind the analysis rather than stopping at advice.
Background
Background and perspective
My physics background lets me work across the cost and energy layers. Earlier work in policy and international affairs helps me read the institutional and political consequences behind the numbers. A postgraduate diploma in Artificial Intelligence with Machine Learning from Humber College connects that background to the systems I build today.