David Tweedle

David Tweedle specializes in LLM inference optimization, focusing on accelerating inference for multimodal and large language model workloads on custom hardware at d-Matrix. They design and test custom numerical formats, model compression strategies, and hardware-efficient implementations of nonlinear activation functions. Previously, David held the position of University Lecturer at the University of the West Indies, where they conducted research in number theory, developed and taught various mathematics courses, and mentored graduate students. David is currently pursuing a double major in Pure Mathematics and Combinatorics & Optimization, along with a Ph.D. in Pure Mathematics at the University of Waterloo.