Kerja Sepenuh Masa, Machine Learning Engineer di Talent Link By E2i - Maukerja

Talent Link by e2i is a programme to match candidates to job opportunities offered by e2i’s Industry Partners. Applicable for Singaporeans and Singapore Permanent Residents only.

This job opportunity is from our Industry Partner in the Public Healthcare Industry.

Job Descriptions

• We are looking for a mid-senior MLOps Engineer to join the Artificial Intelligence Development Office - Company's in-house AI team.
• AIDO functions as an end-to-end product team, building and deploying AI solutions that directly support clinical decision-making.
• You will work closely with our technical leads to build out MLOps and platform engineering functions from the ground up.
• You will be involved in tooling decisions, pipeline setup, deployment workflows, and engineering standards as we scale our clinical AI projects into production.
• Support the design and build-out of AIDO's MLOps platform - contributing to tooling choices, deployment standards, and infrastructure decisions
• Build and maintain CI/CD pipelines for model training, validation, containerisation, and deployment
• Set up and maintain monitoring, alerting, and observability solutions covering model performance, data drift, latency, and infrastructure health
• Maintain clear documentation for infrastructure and deployment processes
• Support and drive the deployment of clinical AI products to cloud infrastructure including AWS
• Implement and configure cloud-native tooling suitable for AI workloads in a regulated healthcare environment
• Support the team's progression toward hybrid and cloud-ready infrastructure Engineering Standards & Team Enablement
• Help the department to establish and adopt engineering best practices - code review processes, branching strategies, deployment checklists, etc.
• Work with data scientists to package models as reproducible, production-ready services (REST APIs, batch inference)
• Translate ambiguous requirements into clear technical plans and execute with minimal direction and approach problems systematically