Handling your AI off-chip is a great way to increase performance, which was the topic of the discussion between Axelera's Fabrizio DelMaffeo and Rich Nass of Embedded Computing Design
