You will own the volumetric streaming pipeline — the system responsible for getting a photorealistic volumetric human from a multi-camera capture rig to a headset in under 300ms, at high visual fidelity, across real-world network conditions.
This is a deep, unsolved engineering challenge. You will work at the intersection of media streaming, network transport and 3D geometry compression — designing systems that must be robust, adaptive and fast enough to feel like real presence.
Responsibilities
Design and optimise congestion control and adaptive bitrate algorithms for volumetric streams
Integrate and benchmark video codecs for combined mesh and texture data
Reduce end-to-end latency across constrained and high-throughput networks
Build monitoring, telemetry and diagnostic tooling for production deployments
Research and prototype new streaming approaches — QUIC, SVC, geometry compression
Collaborate with the SDK team to expose streaming controls at the application layer
Requirements
4+ years in real-time networking or media streaming (video, audio or data)
Strong C++ or Rust; comfortable reading network protocol specifications
Deep knowledge of video codec internals — H.264/H.265, AV1 or similar
Experience with adaptive bitrate, packet loss recovery and jitter buffering
Solid understanding of TCP/UDP transport behaviour under varying network conditions
Nice to have
Prior work on WebRTC, GStreamer or FFmpeg-based pipelines
Experience with QUIC or other modern transport protocols
Background in 3D mesh compression or geometry streaming formats
Understanding of XR rendering constraints — frame timing, reprojection, prediction
Familiarity with multicast or anycast delivery for live streaming at scale