Embedded AI Tools Engineer

Job Title

Embedded AI Tools Engineer

Role Summary

Design and implement embedded AI deployment infrastructure and model optimization tools for heterogeneous SoC architectures (DSPs, NPUs, CPUs). Join ADI's Advanced AI tooling team to enable developers and researchers to move models from concept to production on resource-constrained embedded platforms.

Position is based in Wilmington, MA; travel required ~10%.

Experience Level

Mid-level (Experienced). The posting indicates an experienced-level role but does not specify a required number of years.

Responsibilities

Deliver toolchains, workflows, and infrastructure to optimize and deploy AI models on ADI embedded platforms.

• Design, test, implement, and release AI model deployment tools and infrastructure for heterogeneous architectures (DSPs, NPUs, CPUs).
• Build end-to-end workflows spanning model development, optimization, hardware mapping, and deployment to embedded platforms.
• Develop hardware-aware model design and architecture mapping techniques to leverage specific processor features and mixed-precision computation.
• Implement model compilation and optimization pipelines: quantization, pruning, layer fusion, and hardware-specific code generation.
• Prototype agentic/automated workflows for model-hardware co-optimization and adaptive deployment strategies.
• Collaborate with silicon, firmware, and application teams to integrate tooling with ADI hardware and SDKs.
• Maintain CI/CD, build systems, and infrastructure automation for tool releases.
• Profile and optimize runtime performance on NPUs, DSPs, and other accelerators; support developer tooling (debuggers, profilers).

Requirements

Must-have and nice-to-have technical skills derived from the posting.

Must-have:

• Strong embedded systems and computer architecture experience (bare-metal, RTOS, or embedded Linux).
• Experience with end-to-end AI/ML model development, optimization, and deployment on embedded platforms.
• Proficiency in C, C++, and Python; experience with firmware and low-level software development.
• Deep understanding of neural network quantization, pruning, knowledge distillation, and other optimizations for resource-constrained devices.
• Knowledge of neural network accelerators (NPUs, DSPs) and executing neural networks on heterogeneous hardware.
• Familiarity with AI/ML frameworks (TensorFlow, PyTorch) and deployment tools (TensorFlow Lite, ONNX Runtime, TVM, etc.).
• Experience with build systems (CMake, Make, Ninja), CI/CD pipelines, and infrastructure automation.
• Background with ML architectures (CNN, DNN, Transformer) and their embedded implementation.
• Familiarity with developer tooling such as debuggers, profilers, and SDKs.

Nice-to-have:

• Hardware-software co-design experience and custom operator development for specialized hardware.
• Knowledge of neural architecture search (NAS) and automated model optimization techniques.
• Digital signal processing (DSP) algorithm implementation experience.
• Experience with edge AI frameworks (TensorFlow Lite Micro, ONNX, Apache TVM, MLIR) and FPGA development.
• Understanding of compiler optimizations and code generation for embedded AI accelerators.
• Experience with Zephyr RTOS or other open-source RTOS ecosystems and contributing to open-source projects.
• Familiarity with heterogeneous architectures (ARM, RISC-V, DSPs, custom SoCs).

Education Requirements

Not specified.

About the Company

Company: Analog Devices

Headquarters: Norwood, Massachusetts, USA

Analog Devices is a leading global semiconductor company that bridges the physical and digital worlds, enabling breakthroughs at the Intelligent Edge. With a focus on innovation, ADI develops solutions that drive advancements in digitized factories, mobility, and digital healthcare. The company employs around 24,000 people globally and reported revenues exceeding $9 billion in FY24, creating technologies that transform lives across various sectors.