Nishith Chennakeshava

I'm the CTO at loadmaster.ai, where I build reinforcement learning systems that optimise container terminal operations in Europe. The systems I've built are real-time capable, and currently being deployed at real terminals.

The path here wasn't linear. I started in the British Army Reserves as a reconnaissance soldier, an experience that taught me a lot about operating under uncertainty and within tight constraints. From there, I moved into physics research: first at DIFFER in the Netherlands, working on machine learning for fusion plasma simulation, then at the Max Planck Institute and Princeton Plasma Physics Laboratory, optimising stellarator magnetic geometries.

That research foundation led me to a PhD at TU Eindhoven, where I worked on AI-driven medical imaging, specifically using deep learning and reinforcement learning to make ultrasound imaging faster and more autonomous. The work spanned plane wave compounding, intravascular ultrasound, and shear wave elastography, resulting in publications at venues like IEEE ICASSP, IEEE TMI, and Ultrasound in Medicine & Biology. A collaboration on deep learning for COVID-19 detection in lung ultrasound became one of the most-cited papers in its field.

The thread connecting all of it: I'm drawn to hard problems where the gap between research and real-world deployment is wide, and where bridging that gap requires both deep technical understanding and domain expertise. Container terminal logistics is exactly that kind of problem: combinatorially complex, operationally constrained, and consequential in ways that matter to the physical economy.

Outside of work, I enjoy climbing mountains, and learning new skills.