IC3 - MLOps / Platform Engineer

What You'll Do:

Cloud & ML Infrastructure

• Architect and maintain cloud-native ML & GenAI infrastructure on AWS, including managed services such as SageMaker, Bedrock, EKS, EC2, S3, Lambda, API Gateway, RDS, CloudTrail, and CloudWatch.
• Help your team deploy solutions to Kubernetes clusters across cloud and on-premises environments.
• Develop and maintain Infrastructure-as-Code (IaC) using tools such as Terraform and Helm.

Data & Model Pipelines

• Design, build, and manage multi-stage ETL pipelines to support model training and real-time inference workloads.
• Support model development workflows, including experiment tracking, model versioning, and reproducible training runs.
• Collaborate with ML engineers on fine-tuning, evaluation, and deployment of models, including LLMs and GenAI components.

Platform Reliability & Practices

• Implement observability solutions for ML training and inference pipelines (e.g., Weights & Biases or equivalent tooling).
• Establish and enforce platform patterns and engineering best practices across teams.

What It Takes:

Technical

• Proven experience with AWS managed services: SageMaker, Bedrock, EKS, EC2, S3, Lambda, API Gateway, RDS, CloudTrail, and CloudWatch.
• Proven experience with Kubernetes, both on cloud and on-premises.
• Proven experience designing and operating multi-stage ETL pipelines for ML training and inference.
• Proven experience setting up observability for ML models (training and inference), such as with Weights & Biases (W&B).
• Solid understanding of platform engineering best practices and patterns.
• Hands-on experience with Infrastructure-as-Code tools (Terraform, Helm, or similar).

Leadership & Collaboration

• Ability to work closely with ML engineers, backend engineers, and platform stakeholders on shared, cross-functional systems.
• Comfortable establishing and enforcing platform patterns and best practices across teams.
• Clear communicator, able to align infrastructure decisions with the needs of model development and deployment workflows.

Mindset

• Reliability-minded: builds systems that are observable, scalable, and built to last, not just functional.
• Curious about the full ML lifecycle, from data ingestion and training through to production deployment, rather than infrastructure in isolation.
• Pragmatic and standards-driven, with a bias toward reusable platform patterns over one-off solutions.

Nice to Have

• Experience securing managed and self-hosted AI platforms, including ChatUI integrations, MCP servers, and backend services.
• Familiarity with Apache Kafka and event-driven architectures.
• Hands-on experience with Snowflake integrations.
• Experience with ETL-as-Code frameworks such as dbt.
• Hands-on experience with workflow orchestrators such as Prefect or equivalent (e.g., Airflow, Dagster).
• Proven experience with AWS IAM and account management.
• Familiarity with ML frameworks such as PyTorch, Hugging Face, or scikit-learn.

What This Role Is Not:

• This is not a pure DevOps or SRE role. You will work directly with ML systems, training pipelines, and model deployment - not just maintain cloud infrastructure.
• This is not a data engineering role. While you will build and operate data pipelines, the focus is enabling ML training and inference workflows, not analytics or business reporting.