What a difference a year makes.

When Netflix secured WWE Raw, a three-hour weekly live broadcast with more than three decades of history, it marked a turning point.

Raw is chaos in motion: fan-fueled, unpredictable, impossible to rehearse. And now Netflix has to make it flawless.

More than a licensing play. It’s an engineering gauntlet.

Pulling it off means delivering a resilient platform that can handle global, live-grade performance every single week.

Raising the Stakes

Live streaming has, of course, been around for years. But the expectation that a global service can execute hours of live content flawlessly, at scale, every week, that’s new.

Raw sets the new bar for resilience. And from here on, every live product will be judged against that standard, whether it’s financial trading, real-time auctions, or healthcare collaboration.

How Netflix Built for Weekly Live

To go beyond its on-demand stack, Netflix built a dedicated live media and data pipeline, piece by piece:

• Dual ingest with AWS MediaConnect and MediaLive for redundancy at the source.
• Two-second CMAF/HLS segments tuned for speed without overwhelming origin.
• Wall-clock aligned manifests to sync across devices and regions, even with edits.
• Adaptive ladders in AVC and HEVC balancing SD to 4K playback, mindful of constrained devices.
• Open Connect CDN with 18,000 servers in 6,000 locations, embedded in ISPs to slash startup time.
• Full-stack observability with Atlas, Mantis, and Lumen, surfacing drift and sync issues before users ever notice.

For this approach, Netflix couldn’t build on top of an old stack. They’re deliberate adaptive platform choices that now set the standard for how live systems must be built. They had to replatform in a way that provided the most elasticity in the future.

Fragile Points That Remain

Even with this purpose-built pipeline, the weak spots look familiar to anyone running live systems:

• Start-time surges hammer manifest and session services in the opening seconds.
• Origin contention can still trigger missing or delayed segments.
• Device diversity breaks the chain; older smart TVs or shaky networks fold under jitter.
• Regional edits introduce risk, where one mistimed segment can fracture continuity for millions.

In live systems, small upstream cracks multiply quickly. What feels like a hiccup in testing can become a global outage in production.

The 8th Light Take

At 8th Light, we’ve helped teams in healthcare, finance, and entertainment media face and address these same pressures. The lessons learned in these partnerships transcend industries:

• Resilience is the user experience. Stability may be invisible, but stalls and drift are not.
• Redundancy must be designed in. Dual ingest and encoding pipelines can’t be afterthoughts.
• **Manifest rigor matters. **Segment consistency is a discipline, not a last-minute patch.
• **Adaptive tuning should know the content. **Different scenes stress playback differently.
• Observability must span end-to-end. Ingest, origin, and device playback all need eyes.

The New Baseline

Streaming WWE Raw every week on Netflix proves that massive live events can be delivered with consistency. The baseline has shifted.

For engineering leaders, resilience is the product.

The challenge now isn’t feasibility; it’s whether your systems can withstand failure, recover fast, and keep fans, patients, or traders in the flow without noticing the turbulence.

Live is unforgiving. Fan expectations only climb higher.

You know where we’ll be next week.