Many companies today are focused on trying to deliver peak efficiency in machine learning (ML) inference by encouraging customers to move from less efficient traditional processors, to purpose-built accelerators for ML inference.  While this is directionally correct, oftentimes hardware specific solutions are unable to match customers’ performance and efficiency goals. The issue, solving for ‘peak efficiency’ cannot be accomplished by simply throwing a combination of silicon and power at the problem; this is especially true at the edge. In this fireside conversation, Sakya Dasgupta Founder and CEO of EdgeCortix will examine with veteran industry expert Mike Demler, why software is often the critical bottleneck to productionizing edge-AI hardware acceleration solutions.  The conversation will focus on “thinking software-first” as a primary consideration for customers who are looking for orders of magnitude better performance and energy-efficiency at the edge.   Dasgupta and Demler will review and entertain questions around the importance of leveraging ‘hardware and software co-exploration’ to truly deliver optimal performance and cost benefits to customers deploying novel ML inference hardware at the edge.