Description:

In this episode of Reinventing Broadcast, Ben reflects on Week 5 of the Berkeley ExecEd AI Strategy course — a deep dive into robotics, Markov Decision Processes (MDPs), and reward engineering. Co-host HAIley joins him to unpack how reinforcement learning applies to real-world broadcast scenarios, from robotic camera tracking to live sports aerial rigs.

Ben also shares insights from his two assignments, explores levels of robotic reliability in media workflows, and starts laying the groundwork for his capstone project. It’s one of the toughest modules yet — but one filled with practical relevance for the future of AI in production.

🎓 Topics:

– MDPs explained in a media context

– Real use cases: tape retrieval, lens focus, aerial cam rigs

– Reward functions for smooth framing and battery efficiency

– Capstone project networking and ideation

– Upcoming meetups & conferences

Follow the journey at Reinventing Broadcast as Ben builds toward AI consultancy in a rapidly changing media landscape.