Machine Learning Engineer - Ray

Job Title: Machine Learning Engineer - Ray

Job Duration: 12 months contract with a possibility to extend or to go perm!

Salary: S$ 7,000 to 8,500 /monthly

---

Morgan McKinley is looking for a skilled Machine Learning Engineer - Image Gen for a 12-months contract with our client in the tech industry.

As a Machine Learning Engineer within the ML Platform team, you will play a key role in the creation of our new compute layer using Ray. You will help drive the set-up of the infrastructure for Ray on Kubernetes and its integration with the company's existing Data system.

Responsibilities:

• Develop and Deliver High-Quality AI Infrastructure: Collaborate with the Machine Learning Platform team to design and build infrastructure for distributed data processing and model training using Ray. Ensure reproducibility across Kubernetes clusters by leveraging GitOps for cloud infrastructure management.
• Enhance Observability for Ray: Build and integrate monitoring and alerting solutions within observability stack, which includes Datadog, Prometheus, and Grafana. Contribute to the development of runbooks and DevOps documentation.
• Facilitate Ray Adoption Among Data Scientists: Partner with the product team to promote the use of Ray and provide support for users running jobs on Ray clusters.

Qualifications:

• With 3+ years of experience in Machine Learning Engineering and Ray
• Expertise in ML-Ops and Distributed Computing: Strong knowledge of distributed computing frameworks for data processing. Experience with Ray is a plus, though familiarity with frameworks like Dask, Modin, Beam, Horovod, or Deepspeed is also appreciated.
• Proficiency in Python and ML Ecosystems: Solid experience in Python programming and associated machine learning tools.
• Kubernetes and GitOps Expertise: Deep understanding of developing and deploying systems on Kubernetes. Familiarity with GitOps tools like ArgoCD, along with knowledge of Helm and Kustomize, is preferred.
• Strong DevOps Background: Experience with Infrastructure as Code (IaC) tools like Terraform is desirable.