Senior AI Compiler Engineer for Edge AI Accelerators

Rotterdam Innovation City; Netherlands; Remote; EUR 90,000 - 130,000; Full time

Job summary

Innatera seeks a Senior Compiler Engineer to advance our next-generation compiler framework for AI accelerators. You will own the end-to-end compiler development lifecycle and collaborate with cross-functional peers to extract maximum value for customers.

The role requires hands-on Python/C++ programming, experience with PyTorch or TensorFlow, and familiarity with MLIR/TVM. You will optimize performance and energy efficiency while staying current with NN compiler technologies.

Qualifications

• Proficiency in Python and C++ programming.
• Knowledge of neural networks and familiarity with standard model architectures.
• Hands-on experience with popular deep learning frameworks like PyTorch or TensorFlow.
• Hands-on experience with common compiler frameworks for neural networks, such as MLIR or TVM.
• Strong understanding of computer architecture alongside practical experience optimizing performance for targeted hardware.
• Rapid prototyping capabilities for testing new ideas.
• Strong interpersonal and communication skills.

Responsibilities

• Design and implement the compiler for next-generation AI accelerator architectures, enabling new network topologies while maximizing chip performance and energy efficiency.
• Collaborate cross-functionally with SDK team members as well as algorithm, architecture, and hardware teams to gather compiler requirements.
• Provide feedback to hardware engineers to refine future architectural designs based on insights gained from compiler optimizations.
• Stay at the forefront of neural network compiler technologies, introducing innovative solutions and industry best practices to the team.

Skills

Python

C++

Neural networks

Prototype development

Communication skills

Tools

PyTorch

TensorFlow

MLIR

TVM

GCC/LLVM

Job description

Innatera seeks a Senior Compiler Engineer to advance our next-generation compiler framework for AI accelerators. You will own the end-to-end compiler development lifecycle and collaborate with cross-functional peers to extract maximum value for customers.

The role requires hands-on Python/C++ programming, experience with PyTorch or TensorFlow, and familiarity with MLIR/TVM. You will optimize performance and energy efficiency while staying current with NN compiler technologies.