I am a probabilistic programmer and AI engineer. What that means is that I utilize probabilistic simulation and inference techniques such as Markov Chain Monte Carlo, importance resampling, particle filtering, physical intuition, and computer graphics to develop AI systems that represent the details of what they do and do not know in a common sense way.At IBM I worked with researchers and engineers to support basic research in probabilistic programming for Bayesian computer vision and task and motion planning (TAMP), inspired by the remarkable 3D reasoning capabilities that toddlers develop in the first 18 months of life.On the hardware side I developed a project deploying a Visual Question Answering system on the TIAGo research robot. This autonomous system synthesizes large language models (LLMs) and symbolic reasoning in order to answer spoken questions about real-world scenes.I've also implemented advanced deep neural networks in PyTorch for object detection and pose regression from point-cloud data.