Staff ML Engineer: On-Device AI for Real-Time Games

Unity; Mountain View (CA); On-site; USD 180,000 - 280,000; Full time

Job summary

Unity is seeking a Senior Machine Learning Engineer for On-Device & Mobile AI to optimize and deploy state‑of‑the-art multi‑modal models on mobile and constrained hardware. You will own the inference stack from exported checkpoints to shipped features, balancing latency, memory, and power budgets for billions of players.

You will work with WebGPU-targeted runtimes and native options, implement quantization and pruning, and collaborate across platform, research, and engine teams to ensure robust,

Qualifications

• 5+ years in software/ML engineering focusing on on-device or edge inference.
• Experience deploying transformer- and/or diffusion-based models on mobile/embedded hardware.
• Hands-on with at least one major inference runtime and understanding of fusion, memory layout, and scheduling.

Responsibilities

• Own optimization pipeline: model export, graph transformation, operator fusion, memory layout, and hardware tuning.
• Implement quantization, pruning, distillation to meet latency and memory budgets, validating quality.
• Write/tune WebGPU WGSL and native kernels; profile with browser/platform tools and eliminate bottlenecks.
• Collaborate with runtime teams to integrate ML workloads with game engine scheduling and memory management.
• Develop on-device benchmarking and telemetry in CI; ship feature under memory/power constraints.
• Mentor junior/mid engineers through code reviews and design discussions.

Skills

On-device inference

Model optimization

Quantization

WebGPU/WGSL

Profiling tools

Python

Runtime integration

Transformers.js/TensorFlow.js

Tools

ONNX Runtime Web

CoreML

TFLite

ExecuTorch

WebGPU

Job description

Unity is seeking a Senior Machine Learning Engineer for On-Device & Mobile AI to optimize and deploy state‑of‑the-art multi‑modal models on mobile and constrained hardware. You will own the inference stack from exported checkpoints to shipped features, balancing latency, memory, and power budgets for billions of players.

You will work with WebGPU-targeted runtimes and native options, implement quantization and pruning, and collaborate across platform, research, and engine teams to ensure robust,