This five-year, $199,995 project grant from the National Science Foundation's Computer and Information Science and Engineering program aims to develop a co-designed framework of hardware, software, and algorithms enabling extreme-scale machine learning systems for emerging artificial intelligence of things and internet of senses technologies. Specifically, the Saint Louis University team will pursue five research thrusts: developing hardware and compiler approaches for large-scale split learning across devices; designing an adaptive system partitioning workloads between weak embedded devices and edge computing; addressing data and system unreliability through cross-layer algorithms and hardware techniques; enabling secure and privacy-preserving machine learning at scale through co-design; and implementing testbeds to evaluate proposed system designs. This award supports the NSF's mission to advance computing and information sciences through investigator-initiated research.