Position Details

Salary $110K – $151K USD / year

Type Full-Time

Experience senior

Exp. Years Not specified

Education Not specified

Category Software Engineering

About this role

Embedded AI Tooling Engineer at Analog Devices will build the deployment infrastructure and model optimization tooling for embedded AI on ADI's SoCs. You will design end-to-end workflows, develop hardware-aware optimization pipelines, and explore agentic AI approaches for production deployment.

Key Responsibilities

Design and release AI model deployment tools
Build end-to-end workflows from model development to production
Develop hardware-aware model design and mapping techniques
Design model compilation/optimization pipelines (quantization, pruning, layer fusion, code generation)
Prototype agentic AI workflows for autonomous deployment strategies

Technical Overview

Role focuses on end-to-end AI deployment on embedded platforms, with expertise in CNN/DNN/Transformer architectures, quantization, pruning, layer fusion, and hardware-aware optimization. Tooling spans TensorFlow, PyTorch, TensorFlow Lite, ONNX Runtime, TVM, and multi-architecture compilation pipelines using CMake/Make/Ninja on embedded Linux with RTOS support.

Ideal Candidate

The ideal candidate is a senior embedded AI tooling engineer with a strong background in embedded systems, computer architecture, and deploying ML on resource-constrained SoCs. They should be proficient in C/C++, Python, and ML frameworks, and able to design end-to-end deployment and hardware-aware optimization tooling.

Must-Have Skills

Strong embedded systems and computer architecture experienceEnd-to-end AI/ML model development for embedded platformsHardware-aware neural architecture design and optimizationProficiency in CC++PythonExperience with neural network quantizationpruningknowledge distillationFamiliarity with NPUsDSPs and execution on heterogeneous hardwareExperience with TensorFlow and PyTorchDeployment tools: TensorFlow LiteONNX RuntimeTVMExperience with build systems (CMakeMakeNinja)Embedded Linux and RTOS familiarity

Nice-to-Have Skills

Hardware-software co-designNeural Architecture Search (NAS)DSP knowledgeAgentic AI systems / autonomous optimizationEdge AI frameworks (TensorFlow Lite MicroONNXMLIR)

Tools & Platforms

TensorFlowPyTorchTensorFlow LiteONNX RuntimeTVMApache TVMMLIRCMakeMakeNinjaZephyr RTOS

Required Skills

embedded ai tooling engineerembedded systemscomputer architecturecc++pythontensorflowpytorchtensorflow liteonnx runtimetvmapache tvmmlircmakemakeninjartOSzephyr rtOSfirmwareembedded linuxcnntransformer architecturesquantizationpruninghardware acceleratorsNPUsDSPs

Hard Skills

CC++PythonTensorFlowPyTorchTensorFlow LiteONNX RuntimeTVMApache TVMMLIRCMakeMakeNinjaRTOSZephyr RTOSfirmwareembedded LinuxCNNTransformer architecturesquantizationpruninghardware acceleratorsNPUsDSPs

Soft Skills

strong communication skillsteam collaborationproblem solvingdocumentationself-motivatedleadershipability to translate requirements into toolingstakeholder communication

Industry & Role

Industry Manufacturing

Job Function Develop embedded AI deployment infrastructure and hardware-aware tooling for ADI SoCs

Role Subtype Embedded Engineer

Tech Domains Linux, Python, TensorFlow, PyTorch, TensorFlow Lite, SQL / PostgreSQL

Keywords for Your Resume

embedded ai tooling engineerembedded systemscomputer architectureCC++PythonTensorFlowPyTorchTensorFlow LiteONNX RuntimeTVMApache TVMMLIRCMakeMakeNinjaRTOSZephyr RTOSfirmwareembedded LinuxCNNTransformer architecturesquantizationpruninghardware acceleratorsNPUsDSPs

Deal Breakers

Lack of embedded systems and computer architecture experience, No experience with TensorFlow or PyTorch for embedded deployment, Not proficient in C, C++, Python, No experience with hardware accelerators (NPUs/DSPs)

Apply for this Position →

Get matched to jobs like this

Luna finds roles that fit your skills and career goals — no endless scrolling required.

Create a Free Profile