Skyrim.AI needed to deliver real-time, broadcast-quality spatial media for sports, but volumetric capture from dozens of cameras created interoperability, reconstruction, networking, and data-throughput challenges that traditional broadcast stacks could not handle efficiently.
Reported impact
Skyrim.AI cut development time by a full year, lowered infrastructure costs by replacing specialized hardware with software-defined workflows, and created a platform that supports new revenue streams, richer content production, and personalized sports experiences for global audiences.
Skyrim.AI built its Atlas platform as a software-defined spatial media stack on NVIDIA networking, capture, and inference technologies. Live camera data is sent directly into GPUs with GPUDirect and NVIDIA Rivermax using SMPTE 2110 low-latency transport and PTP synchronization, while NVIDIA Quantum InfiniBand, NVLink, and GPU clustering scale 3D reconstruction across distributed systems. For AI services, the company uses NIM microservices, Maxine NIM, and Instant NeRF-style workflows together with proprietary radiance-field models to reconstruct and render spatial scenes in real time. RTX 6000 Ada GPUs handle camera-to-GPU capture and current inference, while L40S and H100-class GPUs support rendering and future large-scale training. This architecture allows Atlas to deliver broadcast and in-venue experiences from the same spatial capture pipeline while keeping the platform adaptable for OTT, fantasy sports, and interactive fan experiences.
More than two dozen 8K cameras stream synchronized data through Rivermax and GPUDirect into GPUs, reconstruction and radiance-field pipelines process the volumetric data across NVLink and InfiniBand-connected clusters, and Atlas delivers broadcast, venue, and interactive spatial outputs to downstream media experiences.
By leveraging standardized solutions like Rivermax, development time has been significantly reduced by a full year.
Delivering real-time, broadcast-quality volumetric video involves moving over a terabit of data from more than two dozen 8K cameras at 120 frames per second via SMPTE 2110 directly into GPUs with GPU Direct.
NVIDIA's AI and GPU technologies, including NVIDIA NIM™ microservices, NVIDIA Quantum InfiniBand networking, and NVIDIA NVLink™, have enabled Skyrim.AI to build a scalable, high-performance infrastructure with real-time, broadcast-quality spatial media integration.
NVIDIA® GPUDirect® enables efficient data transfer from cameras to GPUs, bypassing the CPU.
NVIDIA® Rivermax® to enhance the capture pipeline, significantly reducing development time and costs.
Review how this reference was scored and trace each normalized claim back to its documented source.
Scoring methodology
Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.
Traceability
Field coverage shows which facts each source supports. Dates are shown only when documented by the source.
Related
NVIDIA Sample Sep 14, 2026 Pinterest Pinterest needed to serve vision-language AI for shopping and discovery at internet scale, but multimodal requests involving thousands of images created prefill-heavy inference, memory pressure, and routing challenges… 14,000 NVIDIA GPUs AI Configurator · Dynamo Planner · NVIDIA Blackwell B200 GPUs NVIDIA Sample Aug 25, 2026 Neuron Systems Neuron Systems needed an always-on AI sports newsroom that could analyze every play, sustain multi-agent reasoning during long live broadcasts, and generate multilingual commentary without the latency, rate limits, and… 1 million audited AI traces NVIDIA CUDA · NVIDIA H100 GPUs NVIDIA Sample Aug 25, 2026 PYLER PYLER needed to verify brand safety across millions of videos per day, but keyword and rule-based moderation could not provide the semantic precision, throughput, or operating cost profile required for real-time… 10X reduction in operational costs NVIDIA CUDA · NVIDIA DGX Platform · NVIDIA Mission Control 