Unity
Staff Machine Learning Engineer
Remote, California, USA +2 · Staff · Full-time · Posted Jul 15, 2026
- Location
- Remote, California, USA
San Francisco, CA, USA
Mountain View, CA, USA - Experience
- 5+ years in software/ML engineering.
- Visa sponsorship
- Work visa/immigration sponsorship is not available.
About the job
Optimizes and integrates multimodal AI models for fast, reliable on-device use in Unity games across mobile and desktop hardware.
What you'll do
- Optimize model export, quantization, kernels, and runtime performance to meet latency, memory, power, and quality budgets.
- Integrate inference runtimes with the game engine, including scheduling, memory pooling, benchmarking, and telemetry.
- Partner with research and platform teams to productionize models, inform model design, and improve engineering practices.
What they ask for
- 5+ years in software/ML engineering, including meaningful on-device/edge inference or real-time performance-critical work.
- Shipped production deployments of transformer- and/or diffusion-based models on mobile, desktop, or embedded hardware.
- Hands-on experience with a major inference runtime and working knowledge of operator fusion, memory layout, and runtime scheduling.
- GPU/compute API and profiling experience; working knowledge of model optimization techniques and strong Python skills.
Nice to have
- WebGPU deployment and WGSL shader experience; game-engine or real-time-graphics background.
- Experience with on-device generative pipelines, inference frameworks, benchmarking infrastructure, compiler stacks, or runtime integration in C++/Objective-C/Swift.
Pay
218 400.00 USD - 283 900. 00 USD (base salary)
Tools
WebGPU, WGSL, ONNX Runtime Web, Transformers.js, WebLLM, TensorFlow.js, CoreML, TFLite
AI focus
Deploys and accelerates transformer, diffusion, vision-language, and other multimodal generative models directly on devices, including model optimization and productionization.
Model-written summary · the original posting is authoritative