LLM Scaling & Distributed Systems Engineer
Posted 2
views
Summary
Orchestrates large-scale LLM pre-training runs in Albury, NSW — designing and managing GPU clusters, automating month-long training campaigns, and optimizing distributed infrastructure for reliable throughput. Core tech: 3D parallelism (Data/Tensor/Pipeline), SLURM/Kubernetes GPU clusters, C++, CUDA.
Hyphen Connect is seeking a highly skilled LLM Pre-training & Distributed Engineer to orchestrate large-scale training runs and optimize distributed infrastructure in Albury, NSW, AU. You will design and manage GPU clusters, provide robust automation for month-long training campaigns, and ensure reliable throughput. The role demands deep expertise in 3D parallelism (Data, Tensor, Pipeline), experience with SLURM or Kubernetes GPU clusters, and a strong systems background in C++, CUDA, and J-188…