Marlowe Seminar Series: Scaling Research with GPUs
The Marlowe team hosted the first event in a new seminar series focused on GPU-accelerated research at Stanford. Presentations covered multi-node training best practices, InfiniBand optimization, and case studies from Cycle 4 researchers who scaled their workflows from single-GPU to cluster-wide training.
The series will continue monthly, with upcoming talks on distributed training frameworks (PyTorch FSDP, DeepSpeed) and efficient data loading strategies.