My background is in physics, and that shapes the way I approach data. Studying physics means working constantly with measurements that are noisy, incomplete or ambiguous, and learning to judge carefully what a result actually supports - and what it does not.
I bring that mindset to data work. I start where many projects stall: structuring, cleaning and storing information so that it becomes a dependable foundation rather than a scattered collection of spreadsheets. From there I build visualisations that make patterns visible and genuinely support decisions. Where questions become complex, I apply machine learning and neural networks - though only where they earn their place. Often a well-chosen statistical model answers the question faster and far more transparently.
In every case I keep results explainable and honest about their uncertainty, so you know not only what the data says, but how far it can be trusted.