Ajay
Kannan.

I grew up in Chennai, where I first fell in love with building things — wiring Arduino boards, training accelerometers to read sign language gestures, and leading an IoT research group before I had a degree to show for it.

That curiosity carried me to two summers at Microsoft, where I shipped sensor-driven ML for agriculture and helped integrate Cortana with a shopping platform — my first taste of building at scale, and of the gap between a model that works in a notebook and one that works in production.

I crossed continents for grad school at Arizona State, researching few-shot learning for immunology — a problem where data is scarce and the stakes are real. That tension between limited signal and high consequence shaped how I think about ML systems today.

Now at Pacific Gas and Electric in the Bay Area, I design the data platform and ML infrastructure behind utility operations — pipelines, anomaly detection, observability, and increasingly, generative AI that has to earn its place in production. The work is less glamorous than a chatbot demo, but reliability is the product.

The journey from Chennai to Santa Clara was not a straight line. It was an odyssey — shaped by curiosity, cross-cultural leaps, and the stubborn belief that the most interesting problems live where engineering meets the real world.