The Food and Drug Administration’s Office of Product Quality Research requires a machine learning and computational statistics platform to detect and classify protein aggregates in biosimilar drug products, supporting a feasibility study for quality assessment and surveillance. The platform must generate morphological fingerprints, differentiate particles by stress type and product, and analyze subvisible particles using neural network-based metric learning, with compatibility for Flow Imaging and Backgrounded Membrane Imaging data. It must also provide quantitative data, employ statistical tools like the Kolmogorov-Smirnov test, and have demonstrable industry experience with prior publications in classifying particle images. The contract will be awarded as a fixed-price contract based on the lowest price technically acceptable method, with evaluation factors including total price and technical conformance. No specific timelines, budget amounts, or performance location are mentioned.