Software Engineer Iii, Speech Production, Infrastructure

Google Google · Big Tech · Mountain View, CA +1

The role focuses on building and maintaining the serving infrastructure for Automatic Speech Recognition (ASR) and Text-to-Speech (TTS) at Google. This involves developing and optimizing high-throughput, low-latency systems that manage bidirectionally-streaming data for compute-intensive models, ensuring reliability and efficiency for various Google products.

What you'd actually do

  1. Write product or system development code.
  2. Deliver the engineering milestones for speech production infrastructure, ensuring high reliability and efficiency.
  3. Profile and optimize serving infrastructure to improve utilization and latency, particularly for compute-intensive models.
  4. Contribute to existing documentation or educational content and adapt content based on product/program updates and user feedback.
  5. Triage and debug complex system issues in production, analyzing the impact on service operations to ensure stability for partners like Search, Assistant and YouTube, among others.

Skills

Required

  • C++
  • Python
  • large-scale infrastructure
  • distributed systems
  • networks

Nice to have

  • data structures
  • algorithms
  • accessible technologies
  • ASR
  • NLU
  • TTS

What the JD emphasized

  • high-throughput, low-latency streams
  • compute-intensive models

Other signals

  • serving infrastructure
  • high-throughput, low-latency streams
  • compute-intensive models
  • ASR
  • TTS