The CTO Podcast with Fexingo · 2026-07-11 · 10 min
Episode 105 of The CTO Podcast dives into Netflix's massive re-architecture of its video encoding pipeline to support 4K HDR streaming at scale. Lucas and Luna break down how Netflix moved from a one-size-fits-all encoding approach to a per-title optimization model that saved 20% bandwidth while improving quality. They discuss the technical trade-offs - per-scene encoding decisions, the shift from CPU to GPU-based processing, and the operational challenge of encoding over 10,000 titles daily. The conversation also explores how this architecture paved the way for adaptive bitrate streaming and the open-source tools Netflix contributed to the community. A behind-the-scenes look at the engineering that powers your favorite shows. #Netflix #VideoEncoding #4KHDR #StreamingArchitecture #CTO #TechLeadership #PerTitleEncoding #GPUEncoding #AV1 #OpenSource #Business #Technology #FexingoBusiness #BusinessPodcast #EngineeringOrg #MediaEngineering #Codec #BandwidthOptimization Keep every episode free: buymeacoffee.com/fexingo
How Datadog Scaled Engineering Without Burning Out82 / 100How Slack Scaled Engineering Without Adding HeadcountHow Atlassian Tamed Technical Debt With Architecture ContractsHow Atlassian Tamed Technical Debt With Architecture ContractsHow GitHub Tamed Merge Conflict Chaos