Founded by Joseph Ward and Saif Al-Ibadi, SmallSpark is dedicated to radically reducing the power consumed by machine learning models to enable their widespread, real-world adoption.
Rather than throwing ever-larger amounts of compute and energy at the problem, we focus on optimising algorithms and hardware to extract maximum performance from minimal power. This approach allows our partners to deploy advanced solutions at scale—without the sky-high electricity bills or the need for monstrous server racks.
By championing on-device processing, we empower organisations to move away from relying exclusively on cloud providers—dramatically lowering costs, cutting out latency, and giving them greater ownership of their data.
Our solutions enable businesses to process information at the edge, tapping into the cloud only as needed. Backed by our team’s diverse expertise in machine learning, software architecture, semiconductors, and electronics, we’re committed to becoming a global leader in the development of ultra-low-power processors for intelligent edge applications. From predictive maintenance to human–machine interfacing, we’re on a mission to drive truly pervasive, cost-effective machine learning that integrates seamlessly with existing workflows.