Machine Learning Engineer (Embedded) (Redmond)
Embedded Machine Learning Software Engineer Location: Fully Onsite at Redmond, WA Contract Duration: 12 Months Pay Rate: $90/hour on W2 Benefits : Health, Dental, Visio, 401K, PTO Note: This is a W-2-only position. We do not offer visa sponsorship and cannot accept H-1B, F-1/STEM OPT, 1099, C2C, or C2H arrangements. Our client's team brings together world-class researchers, developers, and engineers to develop technologies that will shape the future of augmented and virtual reality. The team works across optics, displays, computer vision, audio, graphics, haptics, and advanced perception to enable breakthrough AR glasses, immersive experiences, and next-generation wearable devices. The Audio Research team focuses on advancing human hearing and auditory machine intelligence to improve communication and connection. The team develops and matures audio technologies for AR glasses and wearables, translating early-stage research into innovative capabilities for future products. Job Overview We are seeking a Embedded Machine Learning Software Engineer to bridge the gap between machine learning research and real-world, on-device deployment. In this role, you will collaborate closely with research scientists to transform advanced ML models into efficient, real-time solutions that run on embedded hardware under strict compute, memory, latency, and power constraints. The immediate focus is optimizing and deploying speech extraction models for on-device inference, with opportunities to contribute to a broader range of audio and multimodal machine learning initiatives. The ideal candidate combines hands-on experience training and adapting models using PyTorch or similar deep learning frameworks with strong C/C++ software engineering skills and practical experience optimizing ML inference on embedded or resource-constrained devices. You will take ownership of the deployment process, from model adaptation and optimization through hardware integration, profiling, and performance tuning. This is an opportunity to work alongside leading researchers, solve challenging on-device ML problems, and help bring next-generation AR/VR and wearable technologies to life. Key Responsibilities Partner with research scientists to transition machine learning models from research prototypes to efficient, real-time inference on embedded devices. Train, fine-tune, and adapt ML models while accounting for deployment requirements, including latency, memory consumption, compute capacity, and power efficiency. Optimize models for on-device execution using techniques such as quantization, pruning, knowledge distillation, operator fusion, and architecture adjustments. Port, integrate, profile, and debug ML models on embedded platforms, including DSPs, NPUs, microcontrollers, and mobile system-on-chips (SoCs). Identify and resolve performance bottlenecks across model architectures, inference runtimes, and target hardware. Develop clean, maintainable, high-performance C/C++ code for on-device inference pipelines and real-time prototype systems. Build tools and infrastructure that accelerate model evaluation and enable researchers to assess model accuracy and performance on target hardware. Collaborate with hardware, runtime, and toolchain teams to evaluate and integrate existing technologies and deployment solutions. Document findings, share technical expertise, and contribute to engineering discussions within a multidisciplinary team. Navigate evolving research requirements, explore unfamiliar technologies, and develop practical solutions to complex AR/VR engineering challenges. Must Have Qualifications Bachelor's degree in Computer Science, Computer Engineering, Electrical Engineering, or a related field, or equivalent practical experience. Industry experience optimizing and deploying machine learning inference on embedded or resource-constrained devices . Hands-on experience training machine learning models using PyTorch, TensorFlow, or similar deep learning frameworks . 2 years of industry experience developing software in C/C++. Experience profiling, debugging, and optimizing performance across ML models and target hardware. Experience collaborating with cross-functional engineering or research teams. Preferred Qualifications Experience developing audio machine learning solutions, such as speech enhancement, speech extraction, source separation, keyword spotting, or acoustic event detection. Experience with computer vision machine learning , including object detection, tracking, segmentation, or pose estimation. Understanding of audio digital signal processing (DSP) fundamentals, including filtering, short-time Fourier transforms (STFT), filterbanks, beamforming, and real-time audio processing. Familiarity with on-device inference frameworks and deployment toolchains, such as ExecuTorch, TensorFlow Lite/LiteRT, ONNX Runtime, or vendor-specific NPU/DSP SDKs. Experience with model compression and optimization techniques, including quantization-aware training, post-training quantization, pruning, and distillation. Experience implementing hardware-specific optimizations, including SIMD/NEON instructions, DSP intrinsics, and fixed-point arithmetic. Experience developing software for AR/VR, wearable devices, or embedded consumer electronics . Ability to translate research requirements into engineering solutions and collaborate with researchers on novel approaches. Demonstrated experience building software tools, infrastructure, or prototypes that support research initiatives. Typical Day in the Role You will work at the intersection of machine learning research, embedded software, and hardware optimization. A typical day may involve evaluating a new model with research scientists, identifying deployment constraints, adapting the model for a target device, implementing or integrating the inference pipeline, and profiling performance to achieve real-time execution. You will independently develop approaches to technical challenges, make measurable progress, communicate risks proactively, and contribute to team-wide engineering decisions. Your work will directly support the transition of advanced research into functional hardware prototypes. Pursuant to the California Fair Chance Act, Los Angeles County Fair Chance Ordinance for Employers, Los Angeles Fair Chance Initiative for Hiring Ordinance, and San Francisco Fair Chance Ordinance, qualified applicants will be considered for assignment with arrest and conviction records. Criminal history may have a direct, adverse, and negative relationship with some of the material job duties of this position. These include the duties and responsibilities listed above, as well as the abilities to adhere to company policies, exercise sound judgment, effectively manage stress and work safely and respectfully with others, exhibit trustworthiness, meet client expectations, standards, and accompanying requirements, and safeguard business operations and company reputation. TMMT
