When a major live event is only hours away, “we couldn’t get the GPU instance” is not an acceptable outcome.

Cloud capacity constraints are forcing video teams to reconsider how they approach live transcoding, infrastructure availability, operational complexity, and the true cost of delivering IP video at scale.

In this episode, we sit down with Tim Baldwin, VP of Product Management at Zixi, to explore how a company widely known for reliable video transport evolved into a complete live IP video platform.

We start with Zixi’s roots in transporting broadcast-quality video over unmanaged networks. From there, we examine how Zen Master orchestration, observability, automation, and workflow management help teams operate large fleets of live streams without turning every deployment into a custom engineering project.

Tim explains why protocol-agnostic architecture matters in real customer environments. Depending on who controls each endpoint, a workflow may involve the Zixi Protocol, SRT, RIST, RTP, RTMP, or several of them at once. Supporting that reality allows broadcasters, OTT providers, and technology partners to build around their existing infrastructure and business relationships.

We also get practical about the operational demands of modern streaming workflows, including:

• Normalizing hundreds of incoming streams 
• SCTE-35 conditioning and management 
• Market switching, regionalization, and blackout requirements 
• Dynamic pre-roll, mid-roll, and post-roll insertion 
• HTML overlays and graphics 
• Seamless switching controlled through APIs 
• Rapid deployment for temporary channels and live events 

Tim breaks down the competitive landscape, from internal DIY platforms to traditional broadcast vendors moving legacy products into software and cloud environments.

Then we dig into one of the most urgent infrastructure challenges facing live video teams: transcoding capacity.

What happens when the NVIDIA GPU instance your workflow depends on is unavailable in the required cloud region?

Tim discusses Zixi’s journey from CPU and GPU transcoding to NETINT video processing units (VPUs), including how greater transcoding density can reduce instance counts, infrastructure overhead, and total cost of ownership. We also explore deployments on Akamai Connected Cloud and within Zixi as a Service, where fast onboarding and predictable capacity are critical.

In this episode:

• How Zixi evolved from a media server and transport protocol into a broader live video platform 
• How Zen Master simplifies orchestration, monitoring, and fleet management
• Why protocol-agnostic workflows matter when partners require SRT, RIST, RTP, or RTMP 
• How broadcasters, OTT providers, and virtual MVPDs use the platform 
• Automation for dynamic advertising, graphics, switching, and lightweight production 
• The differences between purpose-built platforms, DIY systems, and legacy broadcast software 
• Why cloud GPU availability is influencing transcoding architecture 
• How NETINT VPUs increase density and reduce the number of required instances 
• Deployment options including licensed software, a SaaS control plane, and Zixi as a Service 
• Pricing models aligned with channels and events rather than protocol usage 
For teams evaluating live streaming architecture, the conversation offers a practical look at reliability, interoperability, cloud availability, automation, and the operational economics behind modern IP video.

Explore Zixi’s live video platform and discover NETINT’s purpose-built video encoding solutions.

Stay tuned for more in-depth insights on video technology, trends, and practical applications. Subscribe to Voices of Video: Inside the Tech for exclusive, hands-on knowledge from the experts. For more resources, visit Voices of Video.