Software Engineer, Acceleration Platform

Google Google · Big Tech · Singapore

Software Engineer on the Acceleration Platform team in Singapore, responsible for designing and implementing AI-native agentic systems to eliminate developer toil. The role involves leading zero-to-one technical initiatives, building scalable frameworks for self-sustaining AI at an enterprise scale, mentoring junior engineers, and ensuring system security and performance. Responsibilities include designing systems where AI agents reason and execute workflows, blending traditional software development with prompt engineering and RAG, creating evaluation pipelines for AI performance, and troubleshooting non-deterministic AI behaviors.

What you'd actually do

  1. Design and build self-sustaining systems where AI agents reason, plan, and execute complex workflows to solve real-world engineering problems.
  2. Blend traditional software development with advanced prompt engineering, Retrieval-Augmented Generation (RAG), and optimization to build highly capable AI applications.
  3. Move beyond standard testing by creating evaluation pipelines to measure AI performance, mitigate, and ensure secure-by-default agent behavior.
  4. Troubleshoot non-deterministic AI behaviors. Investigate model introspection, refine system prompts, and optimize context to keep our self-sustaining systems on track.
  5. Accelerate productivity using AI coding assistants, and collaborate with global teams to discover groundbreaking new ways LLMs can eliminate developer toil.

Skills

Required

  • software programming in Python or C++
  • core ML domain (generative AI, NLP, computer vision, speech/audio, reinforcement learning, recommendation systems, or ML infrastructure)
  • ML infrastructure (model training, model inference, model deployment, model evaluation, optimization, data processing, debugging)

Nice to have

  • data structures and algorithms
  • ability to quickly read and understand complex codebases
  • ability to embrace ambiguity
  • constantly level up debugging and system architecture skills
  • strong core programming skills
  • interact effectively and ethically with Large Language Models (LLMs)
  • agentic workflows
  • developer productivity tools
  • AI safety

What the JD emphasized

  • AI-native agentic systems
  • self-sustaining AI
  • enterprise scale
  • AI engineering excellence
  • AI agents reason, plan, and execute complex workflows
  • prompt engineering
  • Retrieval-Augmented Generation (RAG)
  • evaluation pipelines
  • non-deterministic AI behaviors
  • LLMs

Other signals

  • AI-native agentic systems
  • zero-to-one technical initiatives
  • scalable, resilient frameworks for self-sustaining AI
  • enterprise scale
  • AI engineering excellence
  • design and build self-sustaining systems where AI agents reason, plan, and execute complex workflows