Summary
Anton Bernatskiy is an intermediate QA officer and researcher with 11 years of experience bridging machine learning, robotics, and applied physics. He brings deep academic expertise from a PhD and postdoc focused on autonomous spacecraft shape reconstruction—designing evolvable Boolean circuits and novel fitness-learning algorithms to shrink computation for on-board inference. Comfortable across hardware and software, he has prototyped medical devices, programmed microcontrollers and robotic arms, and implemented system-level Linux tooling early in his career. Now at Subsquid he applies his research-driven problem solving to QA and engineering challenges, marrying rigorous experimental methods with practical product needs. Notably, his work uncovers a bias–variance constraint in digital circuitry for regression tasks, a perspective that informs efficient ML and embedded designs.
11 years of coding experience
9 years of employment as a software developer
Doctor of Philosophy - PhD, Computer Science, Doctor of Philosophy - PhD, Computer Science at University of Vermont
Gymnasium 1 of Neryungri, Yakutia
Lomonosov Moscow State University
Russian, English