Engineers default to the largest, best model. It gets the task done. But token consumption keeps climbing, agents retry, re-read, and loop, and the biggest model isn't always the right one. The usual tradeoff says pick two of speed, accuracy, and cost. This talk covers the holy grail: all three. I'll cover routing strategies between SLMs and frontier models, how to evaluate them, and real-world results from production systems.
David Justus is a technologist working at the intersection of AI and design. He has delivered AI solutions for Nike, L'Oréal, Mayo Clinic, Sunrun, and Panasonic. At Panasonic he focused on generative AI on the edge and efficient software delivery. He is currently founding CTO at ReFwd, an AI forward-deployed-engineering agency, and advises several startups.
