Description

You will manage and optimize distributed systems for large-scale deep learning workloads.

Responsibilities

• Design, deploy, and maintain distributed systems using Kubernetes and Slurm to optimize resource utilization.
• Configure and optimize Multi-GPU and Multi-Node Deep Learning job scheduling.
• Develop complex shell scripts to automate system tasks and reduce manual intervention.
• Monitor system performance and troubleshoot issues within distributed systems and job scheduling processes.
• Collaborate with cross-functional teams to translate project requirements into technical solutions.

Required Skills

• 3-6 years of experience in distributed systems and MLOps.
• Hands-on experience with MLOps tools including MLFlow, Kubeflow, and AutoML.
• Proficiency in Python and Linux systems, specifically Ubuntu.
• Proven experience managing Kubernetes and Slurm environments.
• Experience working with on-prem NVIDIA GPU servers.
• Strong understanding of AI model training, deployment, and Multi-GPU/Multi-Node scheduling.
• Experience with shell scripting for system automation and orchestration.
• Knowledge of logical networks and CI/CD processes.

Preferred Skills

• Understanding of PyTorch, TensorFlow, NLP, or Computer Vision.
• Experience with the NVIDIA ecosystem, including Triton Inference Server and CUDA.

Key Skills

Education

Any Graduate