All jobs
ML Infrastructure & Acceleration Engineer - LLMs
Amsterdam · HybridSenior (5–10 years)Full-timePosted yesterday
About the job
What you'll do
You will own the end-to-end GPU infrastructure for LLM training and inference, ensuring high performance and cost-efficiency. This includes managing clusters, deploying models, and optimizing pipelines at the kernel level.
What they're looking for
Candidates must have 5+ years of experience in ML infrastructure or HPC with hands-on expertise in multi-node GPU clusters and Kubernetes. Proficiency in profiling and accelerating GPU workloads using tools like PyTorch, Triton, or CUDA is essential.
Full description