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 vision AI system that identifies and tracks containers across intermodal yards. The system includes a camera that we call the Brow, which runs vision AI models at the edge to identify inventory in real time. We’ve deployed the Brow to 50 trucks across 3 yards, with plans to expand to 21 total locations. In this talk, we’ll examine key challenges in creating and deploying a vision AI system to power real business outcomes in harsh conditions. We’ll explain how we’ve tackled these challenges, including the design of our camera system, the need for AI at the edge and model training considerations for highly varied environments. We’ll close with a summary of the business impact this system is delivering.