About the job
We are looking for a Machine Learning Engineer with a focus on MLOps, deployment, and performance optimization to help bridge the gap between research and production in the gaming space. You will work cross-functionally with games technical directors, designers, and scientists to ensure that novel AI-driven game concepts can be deployed efficiently across a variety of hardware environments.
Responsibilities
Build and maintain MLOps pipelines: Develop robust CI/CD for ML, model registries, and automated deployment workflows to support rapid iteration.
Optimize for performance: Profile and benchmark models across cloud GPUs and edge devices (e.g., Nsight, PyTorch Profiler) to identify bottlenecks and implement hardware acceleration.
Scale deployment: Design and implement model deployment strategies for both Cloud and Edge environments, ensuring efficient, low-latency execution in game runtimes.
Enhance model efficiency: Apply precision tuning and quantization techniques to meet latency, cost, and memory constraints without significant quality loss.
Collaborate on integration: Work with game engineers to integrate ML models into game engine pipelines and APIs.
Qualifications
Minimum
MLOps & Deployment Expertise: Proven experience with model registries, containerization, and building end-to-end CI/CD pipelines for machine learning. Experience productionizing ML models in the cloud (e.g., AWS and SageMaker endpoints), including scaling, monitoring, and working closely with platform/infra teams.
Hardware Profiling & Acceleration: Experience in profiling and optimizing ML inference on GPUs, with knowledge of CUDA-based runtimes and tools (e.g., Nsight, cuDNN, TensorRT, ONNX Runtime).
Compiler & Runtime Knowledge: Familiarity with graph compiler optimization and tools like MLIR or LLVM.
Framework Proficiency: Extensive experience with deep learning frameworks such as PyTorch, TensorFlow, or JAX.
Strong Software Engineering: Ability to develop high-quality, maintainable code and integrate complex algorithmic solutions into production systems.
Preferred
Hands-on experience deploying ML models on edge, such as iOS or Android devices, including model optimization and hardware-aware inference.
Experience in game development and familiarity with game engines (e.g., Unity, Unreal).
Experience in model distillation, pruning, or other model compression techniques.