Large language models are powerful but often impractical for embedded and on-prem systems due to latency, cost, privacy and memory constraints. Small language models (SLMs)—typically comprised of single-digit billions of parameters or less—offer a deployable alternative but require different expectations […]

In safety-critical computer vision applications such as medical diagnostics and autonomous driving, corrupted training data can be fatal. A famous example involves a dermatology AI that learned to identify rulers rather than tumors, a classic case of data poisoning. This […]

Vision-language models (VLMs) enable “world understanding” tasks—natural language queries, zero-shot reasoning and richer scene context—that go beyond traditional CNN pipelines. But bringing VLMs to the edge is primarily a systems problem: selecting a model you can govern (license and data […]

Many AI medical devices perform perfectly in the lab but fail in the field, for example due to “ungradable” images in the case of retina screening. This talk details the end-to-end engineering journey of a portable, regulatory-cleared retinal camera designed […]

Obtaining suitable training data remains a core challenge in the development of custom computer vision solutions for industrial applications such as those in supply chain and logistics. Industrial data differs fundamentally from common datasets—it is often domain-specific, making it expensive […]

Software-defined vehicles and in-vehicle AI are reshaping automotive compute from dozens of distributed ECUs to centralized and zonal architectures—fundamentally changing how data movement, power delivery, safety and cost are engineered. This talk compares decentralized, centralized and zonal compute approaches and […]

Industrial visual inspection operates at the intersection of perception, decision-making and physical consequences. Despite advances in vision and foundation models, industrial inspection remains fundamentally unsolved: inspected assets are domain-specific and evolving, labeled data is scarce, environments drift and inspection decisions […]

Successful deployment of AI on edge platforms requires system-level strategies that account for critical considerations such as latency bounds, memory behavior and orchestration of heterogeneous hardware cores. In this talk, we present techniques for running deep learning pipelines on edge […]

BNSF is one of four Class I railways in the United States. We maintain 30,000 miles of rail and process 5M+ intermodal shipping containers annually. Maintaining accurate inventory at this scale is a challenge. We’ve developed and deployed an embedded […]