AI/ML Automation Analyst

About the Role

KAUST (King Abdullah University of Science and Technology) is seeking an AI/ML Automation Analyst to join its KSL AI Support Team. This full-time position is based in Makkah, requiring 2-5 years of relevant experience. The Analyst will play a crucial role in developing and maintaining advanced MLOps infrastructure and automation solutions at a supercomputing scale.

Role Context and Objectives

The primary objective of this role is to enhance the efficiency of AI/ML workload deployment and management for researchers. The Analyst will focus on container orchestration and workflow automation, bridging the gap between cutting-edge Kubernetes-based infrastructure and the diverse needs of the research community. This includes contributing to governance, technical enablement, and community development initiatives within KAUST.

MLOps and Infrastructure Development

• Develop and maintain secure, OCI-compliant, and HPC-ready AI/ML and data science software container images.
• Design and implement robust MLOps workflows and pipelines at supercomputing scale.
• Develop and maintain CI/CD pipelines for reproducible infrastructure and workflow deployment.
• Design and deploy APIs for AI/ML services and inference endpoints.
• Implement and manage Kubernetes-based orchestration, including CNI, CSI, and service mesh configurations and optimization.
• Deploy and maintain container registries (Harbor) and model registries (MLFlow, Kubeflow Model Registry).

Governance, Compliance, and User Support

• Provide timely user support via telephone, walk-in, email, and ticketing system for all types of inquiries.
• Maintain high customer service standards in dealing with and responding to user issues and questions.
• Assist in computational readiness reviews for AI research projects.
• Assist in AI model and artifact control reviews to ensure compliance with institutional standards.
• Provide consultation to users on efficient resource usage for AI/ML and MLOps workflows.
• Ensure container images and workflows comply with security policies and best practices.
• Support the implementation of usage monitoring and reporting systems.

Performance Optimization and Benchmarking

• Perform performance debugging and tuning of MLOps and cloud-native workflows.
• Develop and maintain AI/ML and MLOps workload benchmarks for procuring new systems.
• Create and maintain regression testing workloads for existing clusters.
• Deploy and maintain observability and resource monitoring stacks using Prometheus, Grafana, NVIDIA DCGM, and Grafana Loki.
• Contribute to technology evaluation and benchmarking exercises for future infrastructure investments.

Training and Documentation

• Create comprehensive training content for users on MLOps platforms, Kubernetes, and containerization.
• Develop and maintain high-quality user documentation for automation tools and workflows.
• Support the delivery of workshops on CI/CD, container orchestration, and MLOps best practices.
• Contribute to knowledge transfer initiatives within the KAUST research community.
• Provide one-on-one consultation to researchers on efficient use of automation infrastructure.