Join NXP Semiconductors as a Principal AI Engineer (f/m/d) in Munich, Germany

Location: Munich, Germany

Start date: As soon as possible

Duration: Indefinite

About NXP Semiconductors

NXP Semiconductors N.V. (NASDAQ: NXPI) provides High Performance Mixed Signal and Standard Product solutions. Our expertise spans RF, Analog, Power Management, Interface, Security, and Digital Processing. These innovations are utilized in diverse applications including automotive, identification, wireless infrastructure, lighting, industrial, mobile, consumer, and computing. With global operations in over 35 countries, NXP has a workforce of more than 45,000 employees and generates revenue exceeding $10 billion.

Department Overview

As part of the CTO organization, you will be instrumental in driving product innovation at NXP. Our teams serve as the backbone for advanced ideas and solutions in sensing, thinking, connecting, and acting—always in collaboration with internal and external stakeholders to meet customer goals. This role provides the platform to enable NXP’s colleagues, customers, and partners in pioneering breakthroughs in Edge AI that shape our world.

Located in the AI Competence Center (AICC) within NXP's Central Technology Office (CTO), our team collaborates globally on state-of-the-art technology. AICC is the cornerstone for business lines to identify, develop, benchmark, and enable transformative Edge AI solutions on NXP systems.

Key Successful Projects

Notable projects include Software- and Hardware-Aware Neural Architecture Search (NAS), Mixed-Precision Quantization in NAS, and machine learning protection through model watermarking, among others.

Responsibilities

• Support AI/ML use cases from NXP business lines to maximize task performance and minimize resource cost, while identifying future hardware requirements and software enablement features.
• Develop, evaluate, and integrate neural network optimization and deployment methods such as pruning, knowledge distillation, and mixed precision quantization on resource-constrained inference nodes using NXP IP and software environments.
• Investigate integration of these methods into a larger NXP framework, including data-free optimization of neural networks.
• Integrate the above methods and neural networks into larger applications, open-source frameworks, and demonstrators.
• Maintain a methodological approach to track state-of-the-art advancements in relevant fields.
• Interact with NXP application partners in various business lines.

Your Profile and Required Competencies

• University degree: MSc or PhD in Computer Science, Electrical Engineering, or related fields with significant exposure to Machine Learning, Edge AI, and Embedded Systems.
• 9+ years of experience in software engineering with standard development tools (e.g., Git, Bitbucket, unit test-driven development, CI/CD).
• Proven track record in embedded machine learning and Edge AI (publications, hands-on projects, etc.).
• Experience with embedded processors, machine learning accelerators, and embedded software architectures.
• Proficiency with operating systems like GNU/Linux and development boards.

Desired Competencies

• Flexibility with AI frameworks (TensorFlow, PyTorch), preferably using Python and C++ interfaces.
• Experience setting up and maintaining ML development environments (Jupyter, TensorBoard, ClearML, Docker).
• Understanding AI deployment toolchains, portability, and inference engines (CUDA, TensorRT, TFLite, ONNXRT).
• Experience integrating external software libraries and components.
• Strong programming skills in Python, C, C++, and scripting languages on Linux systems.

Additional Competencies for Standout Candidates

• Experience with Deep Learning compiler frameworks (MLIR, TVM, ONNXRT, TFLite).
• ML-DevOps experience.
• Familiarity with distributed compute frameworks (Ray) and shared compute frameworks (Slurm).
• Knowledge of build systems (YOCTO, OpenEmbedded).
• Experience with cross-compilation toolchains for ARM.

Interested candidates can send their reactions to: sebastian.vogel@