I don't know about you, but to me there are few things as interesting as the hardware/software interface: the point where carefully written code meets the messy, physical world of sensors, lenses, and real-time constraints. It's where a clever abstraction either holds up or falls apart the moment a real signal hits it.
That makes Veo a perfect guest. The Copenhagen-based company builds AI-powered cameras that record and analyze sports matches, from grassroots football pitches to professional clubs, and then turn hours of raw footage into something coaches and players can actually use: automatic highlights, player tracking, and match analysis. To get there, they have to capture panoramic video on a custom camera, follow the action without an operator, and crunch an enormous amount of data, reliably and at scale.
My guests sit on both sides of that interface. Anders Hellerup Madsen works close to the metal on the camera itself, on the embedded firmware and the
GStreamer media pipeline that turns raw sensor data into video. Gorm Casper works further up the stack, on the backend that ingests, processes, and analyzes those matches in Rust. Together we talk about where Rust fits across that whole journey, the trade-offs of doing media and computer vision work in a systems language, and what convinced a sports-tech company to bet on Rust for the parts that absolutely cannot fall over.
