Spencer Young

I am an applied mathematician with expertise in probabilistic machine learning and computer vision. Currently, I work as an ML Scientist at Delicious AI, a retail intelligence startup based in Silicon Slopes, Utah, where I lead R&D efforts. My work supports a suite of ML-powered products for CPG brands and retailers, including automated shelf audits, LiDAR-based inventory tracking, and product assortment optimization.

Broadly, my research is motivated by the challenges that arise from deploying AI systems in real-world settings — especially for organizations that lack the vast resources of frontier labs. I’m particularly interested in visual perception, uncertainty quantification, interpretability, and data-efficient learning. Recent publications include a state-of-the-art deep count regression model based on the Double Poisson distribution (ICML 2025), as well as a new metric for evaluating conditional model calibration (ECAI 2025). Coming soon: a paper on Delicious AI’s PriceLens system, which features a novel transformer-based method I developed for learnable product-price attribution from display photos.

Outside of work, I enjoy spending time with my wife and son, playing the piano and organ, cooking, gardening, and hiking. As a Seattle native, I also endure the endless highs and lows that come with being a lifelong Seahawks fan.