Machine Learning Architect Job Description Template

The machine learning architect designs the systems that make machine learning repeatable at an organization: data flow, feature management, training infrastructure, serving and governance. The role is more about platform and process than about individual models, and its value shows in how quickly a new model can go from idea to production without each team rebuilding the same plumbing.

Typical Duties and Responsibilities

• Design the end to end machine learning platform architecture
• Define standards for data access, feature engineering and reuse
• Design training infrastructure including compute and orchestration
• Architect model serving for the organization’s latency and scale needs
• Establish model governance, versioning and lineage requirements
• Evaluate platform tooling and make build or buy decisions
• Design monitoring and observability for deployed models
• Guide teams through adoption of platform capabilities
• Model infrastructure cost and design for sustainable spend
• Document architecture and its trade offs

Education

• Bachelor’s or advanced degree in computer science or a related field

Required Skills and Experience

• 8+ years across machine learning and infrastructure engineering
• Experience designing platforms used by multiple teams
• Deep understanding of training and serving infrastructure
• Knowledge of feature stores, model registries and orchestration tooling
• Ability to make pragmatic build or buy decisions
• Understanding of model governance and lineage requirements
• Strong cost awareness for compute intensive workloads
• Clear architectural documentation

Preferred Qualifications

• Experience with a major cloud machine learning platform
• Background in data engineering as well as machine learning