Our Client, a Mass Media and Entertainment company, is looking for a Senior ML Architect for their Burbank, CA location.
Responsibilities:
- Own the end-to-end ML platform architecture: data ingestion, feature management, training pipelines, model serving, and observability
- Define a phased ML architecture roadmap — from foundational data and tooling layers through scalable orchestration and production serving — and maintain it as a living document
- Make and document key architectural decisions (orchestration approach, serving patterns, feature store design, model registry) with clear rationale, trade-offs, and team alignment
- Ensure the architecture scales as the number of models, data sources, and use cases grows
ML Development Platform Design
- Define the standard ML development platform: environment management, dependency strategy, packaging conventions, and repository standards
- Design the CI/CD architecture for the full ML lifecycle — from DS experimentation through training artifact publishing to production deployment
- Select and specify the right tooling for the DS development lifecycle, with clear guidance on ownership, enablement, and expected usage
- Establish the architectural boundary between experimentation, staging, and production systems
Orchestration & Data Flow Architect
- Design the orchestration strategy and pipeline architecture for ML workflows, selecting the right tooling (e.g., Airflow, Dagster, SageMaker Pipelines) and defining how pipelines should be structured
- Define the data flow from Snowflake through preprocessing, training, and inference — with a clear, visible architectural representation
- Specify structured intermediate storage patterns and data contracts between pipeline stages
- Establish lineage, observability, and monitoring requirements that the MLOps team implements
Deployment & Serving Architecture
- Design an ML-first deployment strategy in AWS: model promotion paths, batch and real-time serving patterns, and infrastructure requirements
- Define the feature store architecture and model registry design to support consistent, traceable model lifecycle management
- Provide architectural guardrails that ensure infrastructure choices (e.g., ECS patterns, SageMaker configurations) serve ML requirements rather than constrain them
- Evaluate and recommend AWS-native and third-party services against the org''s specific ML requirements
Technical Leadership & Cross-Functional Alignment
- Serve as the senior technical authority on ML platform decisions across engineering, data science, and infrastructure teams
- Partner with the ML Ops Lead to ensure architectural designs are implemented to expectations, and that implementation feedback informs architecture evolution
- Work closely with the Data Science team to understand experimentation patterns, tooling pain points, and deployment friction and design for them
- Engage with Data Engineers and Architects to ensure the ML platform is well-integrated with the broader data ecosystem
- Communicate architectural decisions clearly to both technical and non-technical stakeholders
Requirements:
- 8-12+ years of experience in ML Engineering, ML Architecture, or ML Platform/Infrastructure
- Demonstrated experience designing ML platform architecture from scratch in AWS — with specific examples of systems you''ve architected end-to-end
- Deep AWS expertise across services relevant to ML: SageMaker, ECS, Lambda, Step Functions, S3, IAM, RDS, and adjacent data services
- Proven ability to drive technical alignment and set standards across multi-team, multi-repo environments without direct management authority
- Exceptional communication skills — able to make complex architectural trade-offs legible to engineers, data scientists, and business stakeholders alike
- Demonstrated experience designing ML platform architecture from scratch in AWS — with specific examples of systems you''ve architected end-to-end
- Deep AWS expertise across services relevant to ML: SageMaker, ECS, Lambda, Step Functions, S3, IAM, RDS, and adjacent data services
- Strong Python proficiency and fluency with the ML tooling ecosystem in production (e.g., MLflow, orchestration frameworks, feature stores)
- Experience with infrastructure-as-code (Terraform, CloudFormation) and containerization at the architecture and standards level
- Deep understanding of the full ML lifecycle and the specific challenges of bridging DS experimentation with production systems
Why Should You Apply?
- Excellent growth and advancement opportunities
As an equal opportunity employer, ICONMA provides an employment environment that supports and encourages the abilities of all persons without regard to race, color, religion, gender, sexual orientation, gender identity or express, ethnicity, national origin, age, disability status, political affiliation, genetics, marital status, protected veteran status, or any other characteristic protected by federal, state, or local laws.