Software Engineer - AI Platform

We are looking for an experienced Software Engineer specializing in AI platform infrastructure and MLOps services to design, build, and operate the control plane that ties together eBay's entire ML ecosystem. This is a high-impact, full-stack platform role where you will own both core backend MLOps services and the developer-facing AI Hub interface, ensuring every ML practitioner at eBay has reliable, efficient, and intuitive tools to build AI at scale.

You will work on ML Platform Control Plane services (AI Metadata Service, MMS, EMS, Deployminterface, ensuring Experiment Management System (EMS) is built on MLflow, the Model Management System with Python SDK integration, AI Metadata Service, Ray Workspace and JupyterHub notebook infrastructure, distributed tracing and observability across platform services, the AI Hub portal built on React and Node.js, and production monitoring dashboardsall integrated with GitOps-based CI/CD pipelines.

Responsibilities:

• Design and build the ML Platform Control Plane services, including the AI Metadata Service, Management Service, and Deployment Service.
• Develop and operate the Experiment Management System (EMS) built on MLflow for experiment tracking, metrics, artifacts, and lifecycle governance.
• Build and maintain the Model Management System (MMS), including model versioning, lineage tracking, stage transitions, and deployment gating.
• Design and operate the AI Metadata Service to store and serve metadata across experiments, model versions, training runs, datasets, and ML pipelines.
• Build and manage AI Workspace environments, including JupyterHub and Ray Workspaces on Kubernetes.
• Implement distributed tracing and observability across ML Platform services using tools such as OpenTelemetry and Jaeger.
• Design and build the AI Hub portal using React and Node.js .
• Develop and maintain the Python SDK for the AI platform.
• Build and maintain production monitoring and dashboards using Prometheus and Grafana.
• Build and operate CI/CD pipelines for ML workflows and platform services using Argo CD and GitOps-based tooling.
• Collaborate with ML researchers, applied scientists, and data engineers to improve developer workflows and platform usability.
• Improve reliability, scalability, and developer experience across ML Platform control plane services.

Requirements:

• Bachelor's or master's degree in computer science, engineering, or a related field.
• 5+ years of experience building scalable distributed systems or platform engineering solutions.
• Strong programming skills in Python and/or Java.
• Proficiency in TypeScript and JavaScript for React and Node.js development.
• Hands-on experience with MLOps services such as MLflow, Weights and Biases, or equivalent systems.
• Experience designing and operating model management systems with versioning, lineage, and approval workflows.
• Experience building metadata services and scalable data stores for ML platforms.
• Hands-on experience with Jupyter Notebook and Ray Workspace environments.
• Experience implementing distributed tracing across microservices and ML platform components.
• Proficiency with React and Node.js for developer-facing web portals and internal tools.
• Experience designing and building Python SDKs for platform consumption.
• Strong expertise with monitoring and observability tooling such as Prometheus and Grafana.
• Experience with Kubernetes, Docker, and GitOps-based CD tooling such as Argo CD.
• Strong API design, debugging, and performance optimization skills.