Principal Machine Learning Engineer

What you'll do

• Own model export, optimization, kernel tuning, runtime integration, and performance validation within latency, memory, and power budgets.
• Architect the ML-runtime/game-engine integration and supporting model pipelines, fallbacks, telemetry, and device benchmarking.
• Lead and mentor engineers, set performance and code-quality standards, and coordinate with research, platform, and product teams.

What they ask for

• 8+ years in software/ML engineering, including at least 4 years in on-device/edge inference or real-time, performance-critical systems.
• Production deployment of transformer- and/or diffusion-based models on mobile, desktop, or embedded hardware.
• Hands-on WebGPU deployment and WGSL shader tuning, or equivalent deep native GPU/compute API experience with a clear path to WebGPU; experience with a major inference runtime.
• Proficiency in TypeScript/JavaScript, WGSL, and Python; technical leadership experience.

Nice to have

• Experience shipping on-device world-model, neural-rendering, or real-time generative pipelines; game-engine or real-time-graphics background.
• Open-source inference/runtime contributions, WebGPU specification familiarity, compiler-stack familiarity, or on-device benchmarking infrastructure experience.

Pay

Zone A: $250,300 - $325,400 USD base salary; Zone B: $222,600 - $289,400 USD base salary; Zone C: $197,400 - $256,600 USD base salary

Tools

WebGPU, WGSL, ONNX Runtime Web, Transformers.js, WebLLM, TensorFlow.js, CoreML, TFLite

AI focus

Productionize and optimize transformer and diffusion models for on-device generative game experiences, including model compression, runtime integration, and GPU inference acceleration.