Machine Learning Engineer

Bose Corporation, U.S.A•Framingham, MA

1d•Hybrid

About The Position

Bose is seeking a Machine Learning Engineer with expertise in Efficient AI and Edge ML to design, train, optimize, and deploy machine learning models for next-generation audio products. The role involves working closely with embedded, DSP, compiler, hardware, and software engineers to bring AI capabilities to resource-constrained devices. The ideal candidate has strong machine learning fundamentals, practical experience developing efficient models, and an understanding of Edge ML deployment constraints. Deep embedded systems expertise is not required, but the engineer should be able to interpret deployment and integration feedback, identify model-related limitations, and iterate on model architectures and training methods to improve accuracy, latency, memory usage, and power efficiency.

Requirements

M.S. or Ph.D. in Computer Science, Electrical Engineering, Computer Engineering, Machine Learning, or a related field.
Strong foundation in machine learning and deep learning, including model architecture design, training, evaluation, and optimization.
Working knowledge of Efficient ML techniques such as quantization, pruning, knowledge distillation, model compression, and efficient model architectures.
Understanding of how model optimization techniques affect model accuracy, latency, memory footprint, compute requirements, and deployment compatibility.
Proficiency with deep learning frameworks such as PyTorch, TensorFlow, or similar.
Strong Python programming, debugging, and problem-solving skills.
Working knowledge of C/C++ and ability to collaborate effectively with engineers working on embedded and production systems.
Understanding of Edge ML concepts and common resource constraints, including computation, memory, latency, and power.
Ability to analyze model behavior and troubleshoot issues across training, conversion, deployment, and integration workflows.
Strong communication and cross-functional collaboration skills.

Nice To Haves

Hands-on experience with quantization-aware training, post-training quantization, mixed-precision inference, structured or unstructured pruning, operator replacement, or hardware-aware model optimization.
Experience adapting model architectures or training workflows in response to deployment and integration constraints.
Experience deploying machine learning models on edge devices, DSPs, NPUs, mobile platforms, wearable devices, or other resource-constrained systems.
Familiarity with deployment frameworks and toolchains such as TensorFlow Lite, TensorFlow Lite Micro, ONNX Runtime, TVM, MLIR, or vendor-specific SDKs.
Understanding of deployment and performance metrics such as latency, memory footprint, model size, MACs, memory bandwidth, and power consumption.
Experience with audio ML, speech processing, acoustic sensing, or multimodal models involving audio.
Familiarity with LLM- or agent-based engineering workflows, including knowledge retrieval, tool calling, technical documentation analysis, structured data extraction, failure diagnosis, benchmark analysis, or optimization recommendations.
Experience building or evaluating AI-assisted tools that improve ML development, deployment, debugging, or engineering productivity.
Curiosity about the interaction between machine learning models, software toolchains, and hardware platforms.

Responsibilities

Design, train, evaluate, and optimize machine learning models for consumer audio products.
Apply Efficient ML techniques, including quantization, pruning, knowledge distillation, model compression, and efficient model architectures, to improve efficiency while maintaining model quality.
Collaborate with embedded, DSP, compiler, and hardware engineers to deploy and integrate models on resource-constrained platforms, including DSPs, NPUs, microcontrollers, and ARM-based systems.
Analyze deployment and integration feedback, identify model-related issues, and iterate on model architectures and training methodologies.
Adapt models to platform constraints such as unsupported operators, data-type limitations, memory usage, latency, compute requirements, and power consumption.
Profile and evaluate model performance across accuracy, latency, memory footprint, model size, compute efficiency, and power usage.
Develop and maintain repeatable workflows for model training, conversion, compilation, simulation, validation, and performance analysis.
Collaborate across Machine Learning, embedded software, DSP, hardware, compiler, and product engineering teams to deliver production-ready AI solutions.
Evaluate emerging ML technologies and AI-assisted engineering tools that can improve model development, deployment, debugging, and optimization workflows.
Document experiments, technical findings, deployment learnings, and reusable engineering practices.
Mentor interns and co-op engineers.