ML Ops Architect - Azure Platform

This job has now expired please search on the home page to find live IT Jobs.

MLOps Architect Product Dev Functional Lead

Team Coordination

Responsible for coordination and delivery of all Product and Infra work

Organizes and holds weekly check-in meetings for architecture reviews and code reviews

Sets up rhythm for daily syncs

Maintains backlog, including coordinating with project managers to populate features/deliverables and user stories/activities with sufficient information.

Leads pre-sprint planning to review backlog, assign work, and update issues and risks

Identifies dependencies and risks to projects within the area

MLOPs Project Manager

Identify best architecture for given solution in accordance with emerging best-practices in field

Support development of high-level project plans including final deliverables (together with Pillar lead)

Builds and maintains walking deck, strategic communications, architecture diagrams, and all documentation for project

Serves as point of contact for all project-based work

Develops connections with other teams to support project needs

Sets up meetings/sends out notes and action items, etc.

Documentation

Develops and maintains documentation on overall MLOps architecture including pipeline diagrams, component architecture diagrams, and infrastructure diagrams

Release Operations

Responsible for all code releases from development into main

Tests all releases end-to-end before finalizing

Develops and writes release notes to share with operations for change log documentation

Cloud InfrastructureStrong Azure experience is a requirement

Permissions

Understand Azure RBAC roles, responsibilities, etc. for each Azure resource in-scope

Work with SC-Alt account owner to make necessary updates/changes to roles, permissions, resource group configuration in Azure

Develop and Maintain IDWeb access groups as necessary to provide proper permissions to Azure resources

Infra Architecture

Ensure different environments are set up in similar manner to ensure smooth development to production transition

Establish and share best practices for cloud infrastructure set-up

Work with subscription owner(s) to secure additional resources as needed

Maintains service entries in IT systems

Monitoring

Monitor usage and budget; suggest ways to reduce usage and budget if needed

Keep up-to-date on Shadow IT changes and impacts to our resources MLOps Engineering (Shared responsibilities with MLOps Engineers)

New Pipeline Development

Develop new components and orchestrate components into ML pipelines to support product needs

Identify process improvements and make suggestions for implementing

Document all components including inputs, outputs, typical usage

Document pipelines, including production uses and limitations

Pipeline Maintenance

Develop tests to identify issues with typical usage of pipelines

Develop alerts to flag issues with pipelines

Maintain code alongside libraries in-use to ensure secure, up-to-date library versions are in-use

Refactor code to ensure bug-free, streamlined experiences

CI/CD and Release Support

Review all PRs, including suggesting improved implementation techniques and verifying code meets team-established best practices

Create PRs for hot-fix loop-backs to dev

Develop automated CI/CD pipelines to trigger ML pipeline builds when changes are made in pipelines

Dev Support

Serve as a SME on all things AML (accessing data, using notebooks, authentication, running dev tasks including using components, etc.)

Operations Support

Investigate, solve, and deploy bug fixes as needed if bugs in production occur

KT with production operators on pipeline/process changes