Senior Software Engineer (l5) - Cloud Gaming Transport R&d

Netflix Netflix · Big Tech · United States · Remote · Engineering

Senior Software Engineer role focused on designing and implementing ultra-low latency, bi-directional streaming transport protocols for Netflix's Cloud Gaming service. The role involves understanding network path characteristics, refining rate and latency estimation, congestion control, and loss recovery for RTP streams, and collaborating on testing and deployment of new transport stack ideas.

What you'd actually do

  1. designing & implementing highly-interactive, ultra-low latency, bi-directional streams over the public Internet
  2. deep understanding of how the dynamically varying characteristics of internet paths influence the design and implementation of performant cloud gaming transport protocols
  3. refine and improve our understanding of network path characteristics (including in-home WiFi behavior, bufferbloat mitigation on shared access links, and so forth)
  4. refine and import our heuristics for rate & latency estimation, rate adaptation, congestion control, and loss recovery for RTP-based media streams
  5. collaborate with colleagues to implement and deploy new ideas in our server-side transport stack

Skills

Required

  • designing & implementing highly-interactive, ultra-low latency, bi-directional streams over the public Internet
  • deep understanding of network path characteristics
  • rate & latency estimation
  • rate adaptation
  • congestion control
  • loss recovery for RTP-based media streams
  • server-side transport stack implementation
  • lab-based and production (A/B) testing techniques for evaluating service performance
  • network lab environments for controlled data networking experiments
  • modern tools and scripting languages for network traffic capture, test orchestration, post-capture analysis and presentation

Nice to have

  • software development in Linux or closely-related environments
  • hardware encoding pipelines as content sources for RTP streams
  • software development in virtual machine and/or containerized environments
  • partnering with remote engineering or product teams across multiple time zones
  • Masters or Bachelors degree in a relevant science or engineering discipline

What the JD emphasized

  • highly complex transport problems that are open-ended or ambiguous and result in significant business or technical impact
  • designing, implementing and deploying WebRTC-like transport stacks for interactive, ultra low-latency streaming applications
  • how low-latency, two-way media flows over IP networks interact with WiFi paths, bufferbloat, active queue management, and concurrent TCP-like traffic
  • modern strategies for balancing network congestion, goodput, latency and reliability in support of interactive, ultra low-latency streaming
  • network lab environments for controlled data networking experiments exploring large, multi-parameter test spaces
  • A/B testing methodologies for transport stack performance testing in production
  • articulate for diverse audiences the technical and business implications of performance measurement experiments executed both in the lab and in production
  • modern tools and scripting languages used for network traffic capture, test orchestration, post-capture analysis and presentation