I am Clarke Young

I'm a systems builder focused on reshaping how we reconstruct, understand, and interact with 3D worlds.

In recent years, I've led the development of real-time 3D reconstruction systems that are not just research demos — but battle-tested in real-world applications, from autonomous mowing robots to scalable stereo mapping pipelines. My work sits at the intersection of geometry, system design, and deployment.

Currently, I'm building core infrastructure around 3D Gaussian Splatting (3DGS) — aiming for cost-effective, high-fidelity reconstruction at scale. This includes advancing pose optimization, camera–point cloud alignment, and geometry-aware rendering. Rather than tuning existing methods, I focus on architectural redesign, constraint modeling, and efficiency engineering.

My next frontier lies in integrating deep learning, differentiable rendering, and geometry priors to unlock intelligent 3D perception — bridging physical and virtual worlds with precision, performance, and realism.

I'm not here to iterate on the past — I'm here to define what comes next in 3D understanding and generation.