Machine Learning Engineer (AWS)

What You'll Do

• Build and maintain reproducible model training workflows on AWS (SageMaker, S3, Glue, etc.), making retraining, rollback, and experimentation routine rather than heroic.
• Deploy and operate real-time and batch inference services with full CI/CD pipelines, versioning, and safe rollout strategies (canary, shadow, A/B) so changes are deliberate and observable.
• Instrument production models for performance, data drift, latency, and errors — and automate retraining triggers when models drift out of tolerance.
• Maintain model lineage, auditability, and traceability to meet the compliance, governance, and reporting needs of the regulated gaming industry.
• Enforce least-privilege IAM, encryption, and secure data access patterns across the entire ML platform.
• Treat cost as a first-class engineering metric — right-size infrastructure, balance batch vs. real-time workloads, and continually reduce platform spend without sacrificing reliability.
• Collaborate with engineers, data scientists, and product teams to translate business problems into ML solutions, communicate tradeoffs clearly, and iterate based on feedback.
• Continuously explore new AWS services, ML frameworks, and deployment patterns to improve reliability, observability, and developer velocity on the ML platform.

Requirements

• 3+ years of experience in machine learning engineering, MLOps, or a closely related discipline.
• Hands-on experience with AWS ML and data services — SageMaker (training, endpoints, pipelines), S3, Lambda, Step Functions, CloudWatch, MWAA (Apache Airflow).
• Experience working with time series data, including feature engineering, seasonality handling, and temporal train/test splits.
• Strong Python skills and familiarity with common ML frameworks (scikit-learn, PyTorch, XGBoost, or equivalent).
• Experience building and maintaining CI/CD pipelines for ML systems.
• Demonstrated ability to monitor and debug production ML systems — latency, drift, errors, and data quality — and drive issues to root cause.
• Comfort with SQL and working with structured data at scale.
• Able to work collaboratively across teams, assume positive intent, and communicate clearly with both technical and non-technical stakeholders.
• Track record of self-directed learning and technical growth in areas like AWS, ML frameworks, or deployment patterns.

Nice to Have

• Experience in a regulated industry (gaming, finance, healthcare) where auditability, explainability, and compliance are first-class concerns.
• Familiarity with feature stores, model registries, or ML metadata tools (e.g., MLflow, SageMaker Model Registry).
• Experience with infrastructure-as-code (Terraform, CDK, or CloudFormation).
• Exposure to data drift detection libraries or custom drift monitoring implementations.

Success Looks Like

• Production models run reliably with clear, measurable business impact for casino operators.
• Failures are observable, recoverable, and explainable — with logs, metrics, and traces that tell the full story.
• ML systems scale predictably with usage and data volume, without runaway cost.
• The ML platform becomes a trusted, well-understood part of CCT’s product ecosystem — for both internal teams and external customers.