Senior Principal AI Engineer

cerence · Remote

Posted: 17 days ago
Last confirmed live: 3 days ago

What this role involves

This role focuses on designing and operating distributed training systems for large neural networks across GPU clusters. Responsibilities include optimizing multi-node, multi-GPU execution, diagnosing bottlenecks, and improving training stability. The ideal candidate has deep experience with distributed systems, GPU orchestration tools, and training frameworks like PyTorch Distributed, Megatron-LM, and DeepSpeed.

Skills this posting asks for

slurm
kubernetes
ray
runai
nccl
rdma
infiniband
nvlink
pytorch distributed
megatron-lm
deepspeed
activation checkpointing
zero offload
distributed systems
ml systems
gpu clusters
data parallelism
tensor parallelism
pipeline parallelism
hpc

Requirements

Level: senior
Remote policy: remote

Apply without filling the form

Approve this role and the application is completed for you, including a résumé tailored to it. You get a confirmation when it lands, and a credit is only spent when a submission is confirmed.

Claim 25 free applications

Other roles at cerence

All 4 roles at cerence →