Overview
Machine Learning Performance Engineer role at imec. The AI Systems department develops AI compute solutions, from AI models and algorithms to hardware for small-scale edge to large-scale distributed and hybrid architectures. The AI Models group focuses on algorithmic and application research to drive SW-HW-Technology co-optimization and to develop novel ML/Hybrid-AI methods for semiconductor processing and health applications. This position is within the AI Models group based in Leuven, Belgium. We are strengthening the group to build expertise in performance evaluation of Machine Learning workloads and optimization techniques. The candidate will help build a framework to represent distributed AI/ML workloads and decode their execution on different device and network models to enable design of high-performant future exascale systems.
What you will do
AI/ML workload representation toward performance evaluation of exascale systems.
Analyze AI models/workloads to create mini-apps and benchmarks for performance.
Study performance optimization of deep learning workloads for exascale systems.
Explore emerging AI/ML algorithms for custom architectures to enable SW-HW co-design.
Stay up-to-date with the latest AI/ML literature to implement state-of-the-art algorithms.
During the first years, you will have the opportunity to be an integral part of our ongoing EuroHPC DARE project that aims to drive Europe’s independence in high-performance computing and AI.
What we do for you
We offer the opportunity to join one of the world’s premier research centers in nanotechnology at its headquarters in Leuven, Belgium. You’ll become part of a team that makes the impossible possible and helps shape technology for tomorrow’s society.
We are an inclusive employer with an open, multicultural, and informal working environment. imec.academy supports your development to grow technically and personally, with opportunities to take initiative and responsibility.
Your contribution is valued, and imec offers a market-competitive salary with a comprehensive set of fringe benefits.
Who you are
You have a PhD, preferably in Computer Science, Engineering, Mathematics, or equivalent.
Strong background in parallel programming with proven CUDA and Python experience.
Experience with modern Deep Learning frameworks (PyTorch, TensorFlow, Jax).
Experience with distributed training frameworks (e.g., Ray, Dask, PyTorch Lightning).
Strong background in algorithmic optimization techniques such as quantization, pruning, and distributed training.
Experience in software performance analysis and optimization.
Experience optimizing AI for AI Compute Architectures is a plus.
Understanding of performance analysis tools and hardware optimization techniques.
You are a team player with strong communication skills and fluent in English (spoken and written).
Seniority level
Associate
Employment type
Full-time
Job function
Engineering and Information Technology
Industries
Research Services, Nanotechnology Research, and Semiconductor Manufacturing
Referrals increase your chances of interviewing at imec. Sign in to set job alerts for related roles.
#J-18808-Ljbffr