I’m Jim O’Neill. My high school art teacher fundamentally changed the way I see the world. He taught me that what we call lines don’t actually exist, only two planes of color that intersect and give the illusion of one. I’ve been looking for those intersections in data ever since.

I work as a quantitative systems architect applying the scientific method to infrastructure: forming hypotheses about system behavior, designing controlled experiments, building models, and forecasting cost and risk decisions that engineering instinct alone can’t make.

Over the past two decades I’ve done this at scale: a decade building workload analytics for some of the world’s largest MDM deployments at IBM, handling billions of records and trillions of comparisons across Fortune 50 and federal accounts. And now, characterizing the behavior of a global SaaS fleet — digging into API performance, modeling new product workload impacts, building finance-facing deployment projections, and using infrastructure telemetry in ways it was never designed to be used.

My career didn’t start this way. I used to dissolve moon rocks for a living. The discipline required to extract a signal at parts per quadrillion concentrations from a sample you can never replace translates surprisingly well to finding anomalies in a production system you also can’t afford to break.

No lines, and no second chances.

Those two experiences are why this blog exists; from entity and workload analytics to cost intelligence, anomaly detection, and what it looks like when you apply lab science thinking to complex systems.