Shell Remote Sense

Shell Remote Sense was a machine-learning-driven condition monitoring service for industrial equipment. Sensors in the lubricant of operating machinery streamed continuous data, and ML models identified contaminants and signs of degradation, so operators could act before a failure happened.

The product worked, and my job was to grow it: more customers, more markets and more capability.

My role was scaling the product across several dimensions at once. On the commercial side, I worked closely with sales and marketing to bring in new B2B customers: understanding what different industries needed from the platform, translating that into product decisions, and making sure what we were selling matched what we could deliver.

I worked directly with the data scientists who built the ML models, to define what the models could credibly support. I also shaped how confidence and uncertainty were shown to operators, so that they trusted the insight enough to act on it.

On the product side, I led the addition of new monitoring metrics, expanding what the platform could detect and therefore its value proposition to existing and prospective clients. I also worked on the UI, improving how operators interacted with the data and making a technically complex product easier to act on in practice.

Running alongside this was a global launch across five markets: Europe, India, Indonesia, the US and Canada. Each market brought its own operational and regulatory considerations, and coordinating across that many geographies while keeping the product coherent was a significant part of the work.

$400k revenue from six new B2B customers, and a product with meaningfully broader capability and a wider geographic footprint than when I joined.