ML Researcher Intern

At WATER, we are redefining the future of Human Machine Interaction and overall Wellbeing by building the world’s first-ever robotic surfaces—designed to adapt seamlessly to the needs of every individual. Our mission is simple: we believe the world should adapt to humans, not the other way around. This bold vision has never been realized before, and we are pioneering this innovation on a global scale.As we embark on this journey, we aim to create something as iconic as Apple and as revolutionary as Tesla. At WATER, we place the consumer at the center of everything we do, crafting exceptional products powered by the finest engineering and meticulous attention to detail. We are at the intersection of cutting-edge health and wellbeing, fused with the exciting world of robotics—an environment where passion, innovation, and technology converge to change lives.At WATER, we push the edge of adaptive surfaces tech with wild new sensors & analog circuits, the greatest mechanical engineering fused with next-gen ML models, and FW driving bio-robo movements.We’re all about how you think, not what you did before.You’ll get equity here—real skin in the game.We want curious nerds who love tearing stuff apart to find the “why.”You’ll try crazy ideas, fail often, learn fast, and move on.Speak up if your superpower is "Speed & Curiosity."Small team, no BS, just raw, first-principles work.If you can go deep and think at the atomic level, join us.It’s ownership, not a 9-to-5.Specifically for ML Research: We're building a closed loop that learns from the body itself. The reward isn't a label a human typed. It's physiology: heart rate, HRV, respiration, pressure redistribution.You'll build foundational models and an embedded vector space of human body and surface interaction, learned from pressure distribution, BCG, and the metrics we derive from them.Comfortable with modern policy learning and RL? Good.Can you make a model reason over serialized sensor state and emit decisions you'd stake real outcomes on? Even better.Can you close the loop against ground truth, over thousands of sensor points, in real time? That's the kind of wizardry we need.You think in credit assignment, distribution shift, and failure modes, not benchmarks. You'll touch reinforcement learning, on-device perception turning raw sensors into tokens, and a cloud brain that learns the body of a species.Join us, and be part of the team building the future—where exceptional talent meets extraordinary opportunity.Ready to dive in? Let’s chat.