Machine Learning Engineer - Express Scripts Canada

Cigna Healthcare•Mississauga, ON

30d•$115,000 - $125,000•Hybrid

About The Position

Express Scripts Canada (ESC) is the leader in health benefits management. Serving over 12 million members, we help insurance carriers, third party administrators, and the public sector optimize the value of health benefits by linking the talent and professional expertise of our people with leading edge information management systems and technology. Express Scripts Canada is a wholly owned subsidiary of Express Scripts, one of the largest pharmacy benefit management (PBM) companies in North America, part of The Cigna Group (NYSE: CI), a global health company. Together, we deliver innovative, cost-effective solutions that improve access, affordability, and health outcomes for Canadians. Express Scripts Canada is looking for a Machine Learning Engineer to join our Team. The successful candidate will design, build, and operate data pipelines and ML services that detect unusual patterns in healthcare claims and pharmacies. The role blends hands-on data engineering (Oracle + ETL), model development (unsupervised learning), and API serving (Python/FastAPI), with close collaboration across business teams. You’ll ensure the best possible performance, quality, and responsiveness of the applications and pipelines, and help maintain code quality, organization, and automation.

Requirements

5+ years application development experience
3–5 years of professional experience in data science / machine learning engineering with Python, including productionizing data pipelines or ML services
Bachelor’s degree in Computer Science, Data Science, Engineering, Mathematics, or a related field (or equivalent practical experience).
Core programming (Python): Strong with NumPy, pandas, scikit-learn, SciPy (sparse), joblib, CLI (argparse), and data visualization (matplotlib/seaborn).
Modeling (Unsupervised): Practical experience with MiniBatchKMeans/KMeans, DBSCAN, TruncatedSVD/PCA, cluster evaluation (silhouette, CH, DB), and stability/bootstrapping.
Data engineering and Oracle: Writing performant SQL; using SQLAlchemy and oracledb for reads/writes; creating tables/DDL; batch inserts; column/type normalization.
ETL pipelines: Feature engineering over large volumes with chunked processing, environment-aware configs (.env), robust logging, and CSV/DB outputs.
APIs and integration: Batch prediction flows via REST (e.g., httpx client) and schema-compatible exports
Version control and CI/CD: Proficient with Git; familiarity with Jenkins/OpenShift pipelines and on-prem deployment constraints.

Nice To Haves

Experience in Healthcare domain with exposure to Fraud, Waste, and Abuse detection in pharmacy/claims, risk scoring thresholds, and audit support artifacts considered a strong asset.
Experience with on-prem, masked datasets and familiarity with Docker/OpenShift deployment patterns for ML APIs.

Responsibilities

Design & implement ML data pipelines to ingest, engineer, and persist features from Oracle (SQLAlchemy/oracledb), including robust logging, argument-driven CLI tools, and environment-based configuration (.env).
Build and validate unsupervised ML models (e.g., MiniBatchKMeans/KMeans, DBSCAN) with dimensionality reduction (TruncatedSVD/PCA), leveraging chunked processing and sparse matrices for large datasets; evaluate using silhouette/Calinski-Harabasz/Davies-Bouldin and stability checks.
Serve models as REST APIs (ie. FastAPI/Pydantic) with health endpoints, CORS, structured response models, and joblib artifact loading; instrument application logs and operational run scripts.
Orchestrate end-to-end validation (DB → feature engineering → API scoring → curated outputs), writing curated results back to Oracle tables and creating/maintaining schemas and DDL where required.
Collaborate with business teams to interpret clusters/risk buckets, explain top contributing features, and incorporate feedback loops into subsequent runs and outputs.
Support deployment workflows in sandboxed/on-prem environments; participate in CI/CD (e.g., Jenkins/OpenShift pipelines) as part of model operationalization.
Ensure performance, quality, and responsiveness of data pipelines and APIs; help maintain code quality, organization, and automation; perform code reviews and follow Agile ceremonies.
Understand how AI is interpreting the data set and use that understanding to build prompt that lead to expected outcomes
Develop and maintain AI pipelines including data preprocessing, feature extractions, model training, and evaluation