Max Robotics — Global Service Robot Directory + US Certification

About

Vikash K. Mansinghka is a pioneering researcher at the intersection of probabilistic programming, Bayesian inference, and artificial intelligence, with profound contributions to making probabilistic reasoning more flexible, efficient, and broadly applicable. He is perhaps best known for creating Gen, a general-purpose probabilistic programming system that fundamentally advances the field by enabling programmable inference — allowing users to customize and compose inference algorithms rather than relying on rigid, black-box approaches. This landmark 2019 work has accumulated 128 citations and directly addresses critical limitations that previously made probabilistic programming impractical for demanding domains like computer vision and robotics. Mansinghka's broader research agenda centers on bridging theoretical probabilistic modeling with real-world applications. His work on inference metaprogramming introduced powerful new language constructs that give practitioners unprecedented control over inference, while his contributions to goal inference for autonomous agents and 3D pose estimation through probabilistic inverse graphics demonstrate his commitment to embodied intelligence challenges. His AIDE algorithm further advances the field by providing principled tools for evaluating inference accuracy — an often-overlooked but critical practical concern. Spanning Gaussian processes, robotics, and neural scene understanding, Mansinghka's work consistently pushes probabilistic programming toward being a genuinely universal tool for machine intelligence.