Machine Learning / Embedded Inference

Creospan is a growing tech collective of makers, shakers, and problem solvers, offering solutions today that will propel businesses into a better tomorrow. “Tomorrow’s ideas, built today!” In addition to being able to work alongside equally brilliant and motivated developers, our consultants appreciate the opportunity to learn and apply new skills and methodologies to different clients and industries.

******NO C2C/3RD PARTY, LOOKING FOR W2 CANDIDATES ONLY, must be able to work in the US without sponsorship now or in the future***

Job Title: Software Engineer III – Machine Learning / Embedded Inference

Location: Redmond, WA – Onsite

Job Summary

We are seeking a Software Engineer III to develop, optimize, and deploy machine learning models on embedded and resource-constrained devices for next-generation AR/VR and wearable technologies. The ideal candidate will combine hands-on ML model training with strong C/C++ development and experience optimizing real-time inference under strict latency, memory, compute, and power constraints.

Key Responsibilities

• Train, fine-tune, and adapt ML models using PyTorch or TensorFlow.
• Optimize models for embedded deployment using quantization, pruning, distillation, and other performance techniques.
• Port, integrate, profile, and debug ML inference pipelines on embedded hardware, DSPs, NPUs, and mobile SoCs.
• Develop high-performance C/C++ code for real-time on-device inference.
• Collaborate with research scientists and hardware/runtime teams to improve model performance and deployment efficiency.

Required Qualifications

• Bachelor's degree in Computer Science, Computer Engineering, Electrical Engineering, or equivalent experience.
• Industry experience optimizing and running ML inference on embedded or resource-constrained devices.
• Hands-on experience training ML models using PyTorch, TensorFlow, or similar frameworks.
• At least 2 years of industry experience with C/C++.
• Experience profiling and debugging ML performance on target hardware.

Preferred Qualifications

• Experience in audio ML, computer vision, or audio DSP.
• Familiarity with ExecuTorch, TFLite/LiteRT, ONNX Runtime, or vendor-specific DSP/NPU SDKs.
• Experience with model quantization, SIMD/NEON, fixed-point arithmetic, and wearable or AR/VR devices.