Full Job Description
The Opportunity
We are seeking a Senior MLOps Engineer to design, build, and maintain the data and machine learning pipelines that power our AI and analytics platforms. This is a deeply hands-on engineering role covering the full ML lifecycle from data ingestion and transformation through model training, deployment, monitoring, retraining, and rollback. You will bridge data engineering, ML automation, infrastructure, observability, and application deployment to help build scalable, secure, multi-tenant AI infrastructure with a strong focus on reliability, performance, and cost-efficient design.
Key Responsibilities
ML Pipelines & Data Engineering
• Build and automate ML pipelines for data preparation, training, inference, monitoring, and retraining.
• Develop production data flows across Oracle ERP, Snowflake, AWS, and Azure environments.
• Create reusable Kedro pipelines and manage scalable workloads through AWS Batch, EKS, Karpenter, Kueue, and Fargate.
Model Lifecycle & MLOps
• Implement MLflow-based experiment tracking, model versioning, lineage, quality gates, staged promotion, and rollback.
• Capture reproducible run manifests, validate prediction completeness, and support safe partial-run recovery.
• Build observability for run success, completeness, freshness, duration, drift, failures, and infrastructure cost.
Infrastructure & CI/CD
• Provision secure, multi-tenant cloud infrastructure using Terraform or OpenTofu.
• Implement CI/CD workflows using Azure DevOps and GitHub Actions, including testing, scanning, immutable images, and rollback strategies.
• Maintain Dockerized, Kubernetes-native environments using ECR and EKS with appropriately sized compute and memory resources.
Application Deployment & Security
• Deploy secure React and Python ML applications across AWS and Azure using private networking, MFA, RBAC, encryption, and least-privilege access.
• Collaborate with Data Scientists and Product stakeholders to operationalize models, improve performance, and address reliability gaps.
Technical Environment
• Languages & Frameworks: Python (pandas, Polars, boto3, joblib, LightGBM/XGBoost), SQL, JavaScript/React
• Data Engineering: AWS DMS, Athena, Snowflake, Oracle
• Pipeline & Orchestration: Kedro, EventBridge, AWS Batch, EKS, Karpenter, Kueue, Fargate
• MLOps: MLflow, Docker, ECR, Azure DevOps, GitHub Actions
• Infrastructure & Observability: Terraform/OpenTofu, CloudWatch, Prometheus, Grafana, structured logging, drift monitoring
• Cloud & Deployment: AWS (EC2, S3, RDS, Batch, EKS, Fargate, ECR, EventBridge, Lambda), Azure integration
• Security: AWS IAM, Cognito, RBAC, MFA, Secrets Manager, PrivateLink, encryption, network access controls
Qualifications
Essential
• Minimum 5 years of professional experience in ML Engineering, MLOps, or data pipeline development, excluding internships and academic training.
• Proven ability to design, build, and automate production-scale, end-to-end ML pipelines in cloud environments.
• Strong Python and SQL skills, including complex querying against large datasets.
• Hands-on experience integrating Oracle and Snowflake with production ML systems.
• Proficiency with Terraform or equivalent Infrastructure as Code tooling.
• Experience with containerized deployment, CI/CD, and workflow orchestration.
• Experience with Kubernetes/EKS and cloud-native infrastructure.
• Experience with MLflow or equivalent MLOps tooling.
• Understanding of model lifecycle practices: versioning, lineage, quality validation, staged promotion, monitoring, and rollback.
• Experience with AWS and/or Azure cloud platforms.
• Bachelor's or Master's degree in Computer Science, Machine Learning, or a related field.
Preferred
• 7+ years of relevant professional experience.
• Experience across both AWS and Azure environments.
• Familiarity with Kedro, Karpenter, Kueue, or similar orchestration tooling.
• Experience with multi-tenant infrastructure design and cost optimization.
Location: Montreal, Quebec. Hybrid - 2 days per week in office. Remote option available for exceptional candidates.