Embedded ML Engineer: Real-Time On-Device Inference

Crystal Equation Corporation; Redmond (WA); On-site; USD 112,000 - 118,000; Full time

Benefits offered by this job

On-site role in Redmond

Job summary

Crystal Equation Corporation is hiring an ML Software Engineer III to push state-of-the-art models from lab prototypes to real-time embedded inference. You will train, optimize, port, and run models on constrained hardware for wearables and AR/VR devices, balancing quantization and architecture trade-offs.

The role requires deep experience in C/C++, PyTorch or TensorFlow, and achieving efficient performance on DSPs, NPUs, microcontrollers, and mobile SoCs.

Qualifications

• Optimizing and running ML inference on embedded/resource-constrained devices.
• Experience training ML models using PyTorch or TensorFlow.

Responsibilities

• Collaborate with research scientists to port models to real-time on embedded devices.
• Train and adapt models with latency/memory/power constraints in mind.
• Optimize models via quantization, pruning, distillation, and operator fusion.
• Port and profile models on embedded targets (DSPs, NPUs, MCUs, mobile SoCs).
• Write clean, high-performance C/C++ for on-device inference pipelines.

Skills

ML inference

C/C++

PyTorch

TensorFlow

Cross-functional teamwork

Education

BS degree in CS/CE/EE

Tools

ExecuTorch

TFLite/LiteRT

ONNX Runtime

Job description

Crystal Equation Corporation is hiring an ML Software Engineer III to push state-of-the-art models from lab prototypes to real-time embedded inference. You will train, optimize, port, and run models on constrained hardware for wearables and AR/VR devices, balancing quantization and architecture trade-offs.

The role requires deep experience in C/C++, PyTorch or TensorFlow, and achieving efficient performance on DSPs, NPUs, microcontrollers, and mobile SoCs.