“Megatron-LM” jobs in Canada

No matching filters. Try another name.

Results

• 3d

  Senior AI System Engineer - LLM Training System

  Vancouver, BC

  On-siteFull Time

  

  Huawei Technologies Canada Co., Ltd.

  3+ YOECandidates need a Bachelor's, Master's or PhD in a technical discipline and 3+ years in HPC, AI accelerator operators or distributed large-model training. Solid skill in AscendC or CUDA C/C++, C++ and Python, and experience with distributed training frameworks and profiling tools are required.

  AscendC, CANN, Triton, TileLang, MindSpeed, Megatron-LM

  27d

  Researcher - Efficient Architecture for LLMs

  Montréal, QC

  On-siteContract

  

  Huawei Technologies Canada Co., Ltd.

  Candidates need a PhD in computer science or a related field and first-author papers at top ML or NLP conferences. They must know LLM architectures and training tools well, program in Python with PyTorch, and communicate clearly in English.

  Python, PyTorch, Hugging Face Transformers, Hugging Face Accelerate, Megatron-LM, vLLM

  27d

  Senior Researcher - Diffusion-Based Large Language Models

  Montréal, QC

  On-siteFull Time

  

  Huawei Technologies Canada Co., Ltd.

  Candidates need a PhD in computer science, machine learning, statistics, applied mathematics or a related field, plus first-author publications at venues like NeurIPS, ICML, ICLR, ACL, EMNLP or NAACL. Strong Python and PyTorch skills, hands-on diffusion and LLM experience, and fluent English are required.

  PyTorch, Hugging Face Transformers, Hugging Face Diffusers, DeepSpeed, Megatron-LM, vLLM

  27d

  Researcher - LLMs

  Montréal, QC

  On-siteContract

  

  Huawei Technologies Canada Co., Ltd.

  Candidates need a Master's or PhD in computer science or a related field, or equivalent practical experience. They need strong NLP and LLM knowledge, expert Python and PyTorch skills, and experience with LLM training and inference tools. English proficiency is essential.

  Python, PyTorch, Hugging Face Transformers, Accelerate, Megatron-LM, vLLM

• 23d

  AI Systems Engineer – AI Model (Training & Inference)

  Markham, ON

  Full Time

  AI

  AMD, Inc.

  Candidates need hands-on experience shipping AI/ML infrastructure, training and running large models on GPUs, and building GPU kernels. A Bachelor's, Master's or PhD in computer or software engineering or computer science is preferred.

  AMD Instinct MI-series GPUs, RCCL, Kubernetes, PyTorch, vLLM, SGLang

• 28d

  AI Systems Engineer – AI Model (Training & Inference)

  Markham, ON

  Full Time

  AM

  Advanced Micro Devices, Inc

  Candidates need hands-on experience shipping LLMs on real hardware and writing GPU kernels that improved production performance. They also need experience building distributed AI infrastructure, plus a relevant bachelor's, master's or PhD in computer or software engineering or computer science.

  HIP, Triton, MLIR, ROCm, RCCL, Kubernetes

• 29d

  Senior AI Compute Infrastructure Engineer

  Remote in Canada

  RemoteFull Time

  

  Kraken

  5+ YOECandidates need 5+ years of infrastructure engineering experience, including GPU or ML infrastructure in production. They also need Kubernetes and Linux skills, Python, experience with serving frameworks such as vLLM or Triton, and a track record of controlling compute costs.

  Kubernetes, Linux, Python, vLLM, Triton Inference Server, TensorRT

4 companies