Intern (f/m/d) AI/ML Compiler Engineer

TL;DR

Intern (f/m/d) AI/ML Compiler Engineer (MLIR/Edge AI): Developing compiler methods that map neural-network workloads onto heterogeneous edge devices with an accent on AI acceleration, embedded systems, and open-source compiler technologies. Focus on implementing and integrating deployment methods, enabling Neural Processing Units, and solving challenges across machine learning and low-level software development.

Location: Munich, Germany; hybrid setup available

Salary: €2,420 per month

Company

A semiconductor and edge-computing company developing hardware and software solutions for AI-enabled devices.

What you will do

• Implement compiler methods for mapping AI workloads onto NXP edge devices.
• Develop and integrate neural-network deployment solutions for heterogeneous systems.
• Study and apply MLIR and other open-source AI/ML compiler technologies.
• Enable machine-learning acceleration on embedded devices and Neural Processing Units.
• Collaborate with an international compiler team across multiple locations.

Requirements

• Currently pursuing a Master's degree in computer science, computer engineering, or a related field.
• Very strong programming skills in C, C++, or Python.
• Profound knowledge of embedded systems and experience with Linux environments.
• Fundamental knowledge of neural networks and frameworks such as PyTorch or TensorFlow.
• Good English communication skills for collaboration with a multinational team.
• Independent, accurate, critical, and analytical approach to work.

Nice to have

• Experience with AI compiler frameworks such as MLIR.
• Experience with Neural Processing Units.
• Experience with VS Code and Git.

Culture & Benefits

• Flexible working hours and a hybrid work setup.
• 2.5 vacation days per month in case of a voluntary internship.
• Access to an on-site cafeteria.
• Networking initiatives and Employee Resource Groups supporting professional and personal exchange.
• Security-related work may require a conscious and reliable working style because assignments can fall within security certifications.