21 Oct
|
Hyphen Connect
|
Warrnambool
21 Oct
Hyphen Connect
Warrnambool
Hyphen Connect in Warrnambool, VIC, AU seeks a highly skilled LLM Pre-training & Distributed Systems Engineer to orchestrate large-scale GPU training runs and optimize distributed infrastructure.
You will manage 1,000+ GPU clusters with PyTorch, DeepSpeed or Megatron-LM, improve networking (InfiniBand/RDMA), automate checkpointing, and ensure reliable, long-running training jobs. A robust background in C++, CUDA, Python is required.
📌 LLM Training Architect: 1k+ GPU Distributed Infra (Warrnambool)
🏢 Hyphen Connect
📍 Warrnambool