Embedded ML Engineer — Real-Time On-Device Inference

ICONMA; Redmond (WA); On-site; USD 120,000 - 190,000; Full time

Benefits offered by this job

Health benefits

Referral program

Growth opportunities

Job summary

ICONMA is seeking a Software Engineer III for our Redmond, WA client to advance on-device ML across embedded hardware. You will collaborate with researchers to move prototypes into real-time inference and optimize for latency, memory, and power, while delivering robust C/C++ software.

The role centers on training and adapting models, applying quantization and pruning, and porting models to DSPs/NPUs/mobile SoCs. Strong cross‑functional teamwork and ownership are essential.

Qualifications

• BS degree in Computer Science, Computer Engineering, Electrical Engineering, or related field.
• Experience optimizing and running ML inference on embedded or resource-constrained devices.
• Experience training ML models with PyTorch or TensorFlow.
• 2+ years of industry experience writing C/C++.
• Experience profiling and debugging performance across ML models and hardware.
• Experience working in cross-functional teams.
• Preferred: ML for audio, computer vision, and audio DSP.
• Experience with on-device inference frameworks and toolchains.

Responsibilities

• Work with researchers to move ML models from prototypes to real-time embedded inference.
• Train, fine-tune, and adapt models with latency, memory, and power constraints.
• Optimize models for on-device inference using quantization, pruning, distillation, and fusion.
• Port and profile ML models on embedded targets (DSPs, NPUs, MCUs, mobile SoCs).
• Write clean, high-performance C/C++ code for on-device pipelines.
• Build tools to accelerate evaluation of model performance on hardware.
• Coordinate with hardware, runtime, and toolchain teams to integrate systems.
• Collaborate in a multifunctional team and maintain good documentation.
• Learn continuously and tackle AR/VR problems with unfamiliar tech.

Skills

PyTorch

C/C++

Embedded systems

ML inference

Cross-functional teamwork

Education

BS degree in CS/CE/EE

Tools

ExecuTorch

TFLite/LiteRT

ONNX Runtime

NPU/DSP SDKs

Job description

ICONMA is seeking a Software Engineer III for our Redmond, WA client to advance on-device ML across embedded hardware. You will collaborate with researchers to move prototypes into real-time inference and optimize for latency, memory, and power, while delivering robust C/C++ software.

The role centers on training and adapting models, applying quantization and pruning, and porting models to DSPs/NPUs/mobile SoCs. Strong cross‑functional teamwork and ownership are essential.