Principal Machine Learning Engineer (On-Device AI)

Location: Mountain View, California, USA

Base salary: $250,300–$325,400 in Zone A; $222,600–$289,400 in Zone B; $197,400–$256,600 in Zone C. The range reflects annual base salary.

Company

hirify.global develops a leading game engine used to create games, XR and web experiences, and 3D applications across industries including automotive, manufacturing, and healthcare.

What you will do

• Own the end-to-end optimization pipeline for transformer and diffusion models, from export and graph transformation through operator fusion, quantization, pruning, and hardware-specific kernel tuning.
• Develop and optimize WebGPU compute shaders and native GPU kernels across mobile NPUs, mobile GPUs, and desktop GPUs using profiling and frame-capture tools.
• Design the integration between ML inference runtimes and the game engine, including scheduling, threading, memory pooling, zero-copy buffers, and frame-budget management.
• Build model packaging, asset pipelines, device fallbacks, SKU-aware capability tiers, telemetry, and automated on-device benchmarking.
• Partner with research scientists to productionize new architectures and assess efficient inference methods against latency, quality, memory, and power targets.
• Lead and mentor engineers while establishing performance standards, benchmarking practices, code-review guidelines, and regression gates.

Requirements

• 8+ years of software or ML engineering experience, including at least 4 years in on-device, edge-inference, or real-time performance-critical systems.
• Production experience deploying transformer or diffusion models on mobile, desktop, or embedded hardware.
• Hands-on WebGPU experience, including ONNX Runtime Web, Transformers.js, WebLLM, or TensorFlow.js, plus WGSL shader development; equivalent native GPU or compute API expertise may be considered.
• Deep experience with an inference runtime such as ONNX Runtime, CoreML, TFLite, or ExecuTorch, including operator fusion, memory layout, and runtime scheduling.
• Strong knowledge of GPU or compute APIs, model optimization, target hardware, TypeScript/JavaScript, WGSL, and Python.
• Professional English proficiency is required for regular communication with colleagues and partners worldwide.

Nice to have

• Experience with neural rendering, world models, real-time generative pipelines, or on-device diffusion.
• Game-engine or real-time graphics experience with hirify.global, Unreal, or custom engines.
• Contributions to ML inference, runtime, GPU, or WebGPU open-source projects.
• Familiarity with WebGPU compute features, compiler stacks such as MLIR, TVM, IREE, or XLA, and device-farm benchmarking infrastructure.

Culture & Benefits

• Health, life, and disability insurance, with offerings varying by country and employment status.
• Employee stock ownership and retirement or pension plans.
• Vacation, personal days, parental leave, and family-care programs.
• Mental health support, employee resource groups, training and development programs, and an employee assistance program.
• Commute subsidy, office snacks, and volunteering and donation matching programs.