Software Engineer, Genai Silicon Automation, Deepmind

Google Google · Big Tech · Paris, France

Software Engineer role at Google DeepMind focused on building AI agents and systems for testing and evaluation. Responsibilities include creating agent testing systems, developing test problems in simulators, visualizing results, building leaderboards, testing algorithms on robots, and applying formal methods, constraint programming, MLIR, differentiable programming, generative ML, and reinforcement learning to design computing systems. The role involves working with cutting-edge AI agents and collaborating with global teams.

What you'd actually do

  1. Write product or system development code.
  2. Participate in, or lead design reviews with peers and stakeholders to decide amongst available technologies.
  3. Application of formal methods and constraint programming to the verification of complex computing systems.
  4. Perform MLIR-based compiler construction.
  5. Oversee the application of differentiable programming, generative machine learning and reinforcement learning to design and implement domain-specific computing systems.

Skills

Required

  • software development
  • programming languages
  • formal methods
  • constraint programming
  • MLIR
  • differentiable programming
  • generative machine learning
  • reinforcement learning

Nice to have

  • Java
  • C/C++
  • C#
  • Objective-C
  • Python
  • JavaScript
  • PHP
  • Ruby
  • Go
  • machine learning algorithms and tools
  • PyTorch
  • JAX
  • TensorFlow
  • artificial intelligence
  • deep learning
  • LLMs
  • natural language processing
  • compiler infrastructure
  • LLVM
  • GCC
  • programming language formalization
  • low level ML accelerator programming
  • compiler
  • close to hardware performance programming

What the JD emphasized

  • general-purpose learning agent
  • cutting edge AI agents
  • machine learning
  • Neuroscience
  • formal methods
  • constraint programming
  • MLIR-based compiler construction
  • differentiable programming
  • generative machine learning
  • reinforcement learning
  • low level ML accelerator programming
  • compiler or other close to hardware performance programming

Other signals

  • AI agents
  • ML
  • Neuroscience
  • generative machine learning
  • reinforcement learning