Staff Software Engineer, Content Safety, Infrastructure

Google Google · Big Tech · Singapore

Staff Software Engineer focused on building and scaling content safety solutions, including agentic workflows and multimodal understanding, for Google's products, with a strong emphasis on ML infrastructure and model optimization.

What you'd actually do

  1. Safeguard business-critical products that represent significant business, market share or strategic value to Google, including GenAI-based experiences both server-side and on-device.
  2. Design, build, maintain and scale content safety solutions (e.g., scalable classifiers, multimodal understanding) to make users’ experiences safer.
  3. Work on projects such as agentic workflows for threat understanding and content moderation.
  4. Facilitate alignment and clarity across teams on goals, outcomes, and timelines. Influence and coach a distributed team of engineers.
  5. Lead the design and implementation of solutions in specialized ML areas, optimize ML infrastructure, and guide the development of model optimization and data processing strategies.

Skills

Required

  • software development
  • software products
  • software design and architecture
  • Speech/audio
  • reinforcement learning
  • ML infrastructure
  • ML design
  • model deployment
  • model evaluation
  • data processing
  • debugging
  • fine tuning

Nice to have

  • data structures and algorithms
  • technical leadership
  • content safety
  • scaling pipelines
  • deploying global-scale systems
  • defensive architecture
  • Responsible AI
  • factuality
  • product policy
  • machine learning
  • large language models (LLMs)
  • transformers
  • activations
  • efficiently train and deploy them at scale

What the JD emphasized

  • content safety
  • agentic workflows
  • multimodal understanding
  • ML infrastructure
  • model optimization
  • data processing strategies
  • Speech/audio
  • reinforcement learning
  • ML design
  • ML infrastructure

Other signals

  • content safety solutions
  • agentic workflows for threat understanding
  • multimodal understanding
  • ML infrastructure
  • model optimization and data processing strategies