Staff Software Engineer, Quality, Google Cloud, Applied AI

Google Google · Big Tech · Sunnyvale, CA +1

Staff Software Engineer focused on building and improving AI agents, specifically a meta-layer of agents that build, test, and refine other agents. The role involves developing multi-modal customer support agents with complex reasoning capabilities, embedding Gemini and Vertex AI into client-facing infrastructure, and establishing engineering benchmarks for automated optimization and AI quality assurance in production systems.

What you'd actually do

  1. Build the core logic for multi-modal customer support agents that execute complex reasoning across sales and support.
  2. Develop automated systems and tools that allow agents to iteratively build, test, and refine other agents.
  3. Architect the pathways that embed Gemini and Vertex AI intelligence directly into client-facing Cloud infrastructure.
  4. Establish engineering benchmarks to replace manual "trial-and-error" testing with automated, high-fidelity optimization.
  5. Take ownership of AI quality for production systems by defining technical metrics aligned with business goals, implementing evaluation frameworks, designing experiments, analyzing loss patterns, and driving improvements through system changes or training data enhancements.

Skills

Required

  • software development
  • Artificial Intelligence
  • Distributed Systems
  • LLMs
  • High Performance Computing

Nice to have

  • multi-agent systems
  • evaluation frameworks for AI quality in production
  • model evaluation
  • context engineering
  • benchmarking
  • testing agentic systems

What the JD emphasized

  • multi-modal customer support agents
  • complex reasoning
  • agents that build, test, and improve other agents
  • Gemini and Vertex AI
  • engineering benchmarks
  • AI quality for production systems
  • evaluation frameworks

Other signals

  • building agents
  • multi-modal intelligence
  • complex reasoning
  • meta-layer of AI
  • suite of agents designed to build, test, and improve other agents
  • Gemini and Vertex AI
  • customer support agents
  • AI quality for production systems
  • evaluation frameworks
  • benchmarks