Software Engineer, Acceleration Platform Team

Google Google · Big Tech · Singapore

Software Engineer on the Acceleration Platform Team building scalable, AI-native agentic systems to automate complex workflows and solve enterprise-scale engineering problems. The role involves leading zero-to-one initiatives, defining AI-first engineering best practices, scaling evaluations and guardrails for AI safety, and building advanced telemetry for debugging and optimizing self-supporting behaviors.

What you'd actually do

  1. Architect agentic ecosystems by leading the design and implementation of highly scalable, fault-tolerant systems where multi-agent networks reason, plan, and execute complex workflows across vast, distributed codebases.
  2. Pioneer AI-first engineering by defining best practices for the team and broader organization. Blend traditional distributed systems architecture with advanced Large Language Model (LLM) orchestration, complex Retrieval Augmented Generation (RAG) pipelines, and optimization.
  3. Scale evaluations and guardrails by establishing a comprehensive technical strategy for AI safety, architecting, automated frameworks that measure performance and enforce security to mitigate across large-scale deployments.
  4. Solve the hardest AI problems through managing the most intricate non-deterministic edge cases. Build advanced telemetry and introspection tooling that allows the entire organization to understand, debug, and optimize self-supporting behavior.

Skills

Required

  • software programming in Python or C++
  • data structures and algorithms
  • implementing core Machine Learning (ML) concepts

Nice to have

  • AI safety
  • enterprise security
  • advanced prompt engineering
  • scalable model evaluation methodologies
  • distributed systems architecture
  • core programming
  • LLM capabilities, limitations, and failure modes
  • designing, deploying, and scaling LLM-backed applications
  • complex RAG systems
  • self-supporting agents in enterprise production environments

What the JD emphasized

  • AI-native agentic systems
  • automate complex workflows
  • enterprise-scale engineering problems
  • AI engineering
  • multi-agent networks reason, plan, and execute complex workflows
  • LLM orchestration
  • Retrieval Augmented Generation (RAG) pipelines
  • AI safety
  • automated frameworks that measure performance and enforce security
  • Solve the hardest AI problems
  • non-deterministic edge cases
  • Build advanced telemetry and introspection tooling
  • LLM-backed applications
  • self-supporting agents in enterprise production environments

Other signals

  • AI-native agentic systems
  • automate complex workflows
  • enterprise-scale engineering problems
  • AI engineering
  • multi-agent networks reason, plan, and execute complex workflows
  • LLM orchestration
  • Retrieval Augmented Generation (RAG) pipelines
  • AI safety
  • automated frameworks that measure performance and enforce security
  • Solve the hardest AI problems
  • non-deterministic edge cases
  • Build advanced telemetry and introspection tooling
  • LLM-backed applications
  • self-supporting agents in enterprise production environments