We're looking for a Kernel Engineer to join our team in Portugal in a remote working mode. In this role, you will design, optimize and maintain high-performance software components that power modern machine learning workloads. You will work on low-level kernel development for accelerators, optimize performance, and collaborate with cross-functional teams to deliver scalable and efficient solutions that enable cutting-edge AI systems.

Responsibilities

• Design, develop and maintain software kernels targeting advanced accelerators such as TPU/GPU
• Optimize performance-critical code using C++, Python and low-level programming techniques
• Implement and test efficient compute kernels for ML frameworks including JAX and PyTorch
• Enhance developer infrastructure by creating high-performance APIs, OSS libraries and documentation
• Collaborate across engineering teams to solve performance bottlenecks and ensure scalable solutions
• Investigate, profile and optimize workloads including attention mechanisms and Mixture of Experts (MoEs)
• Contribute to code design and architecture reviews for robustness and maintainability
• Ensure software quality through thorough testing, debugging and performance analysis

Requirements

• Bachelor’s degree in Computer Science or equivalent practical experience
• Minimum 5 years of experience in software development with C++ or Python
• At least 3 years of experience in testing, maintaining or launching software products
• Minimum 1 year of experience in software design and architecture