Summary
David Peabody is a product and program manager with nine years of experience at the intersection of AI, GPU software, and real-time 3D, currently shaping hipDNN and Composable Kernels roadmaps at AMD. He combines a chemical engineering background with machine learning and software product management training to translate complex ML primitives—GEMM, convolutions and attention—into performant GPU solutions for LLMs, vision and recommender systems. Previously he led client-facing delivery and internal product strategy at Unity and built analytics and ML teams delivering computer vision and LLM solutions in industry. Known for bridging deeply technical subject-matter experts and commercial stakeholders, he excels at turning research-grade kernels into client-ready releases. Based in Calgary, he still leans on his process engineering roots and open-source curiosity—inviting contributions on his GitHub—while driving pragmatic roadmaps that balance performance, usability and business fit.
9 years of coding experience
15 years of employment as a software developer
Specialization, Software Product Management, Specialization, Software Product Management at University of Alberta
Nanodegree, Machine Learning Engineer, Nanodegree, Machine Learning Engineer at Udacity
2.1 BEng hons, Chemical Engineering, 2.1 BEng hons, Chemical Engineering at The University of Sheffield