Primary Purpose
Execute hands-on data engineering, feature engineering, and model development tasks; contribute to deployments and monitoring under senior guidance.
Key Responsibilities
- Data Engineering
- Build and maintain ingestion jobs (batch/near-real-time) from core systems and external sources.
- Implement data validation, quality checks, and lineage tagging.
- Feature Engineering
- Create domain features: transaction velocity, rolling aggregates, delinquency buckets, KYC/AML risk indicators, digital behaviour metrics.
- Register, version, and document features in the feature store.
- Modelling & Evaluation
- Train baseline and advanced models (logistic regression, gradient boosting, tree ensembles, simple neural nets).
- Handle class imbalance (SMOTE, class weights), cross-validation, and calibration; produce model cards and explainability reports.
- MLOps
- Containerize models, write unit/integration tests, prepare pipelines for CI/CD, and implement inference endpoints/batch scoring.
- Build dashboards for performance and drift; execute retraining under playbooks.
- Documentation & Compliance
- Maintain clean, reviewable code; produce data dictionaries, feature specs, and runbooks.
- Follow security, privacy, and access control policies
Required Skills & Experience
- 1-2 years in applied ML/data engineering, preferably in financial services.
- Proficient in Python and SQL; familiarity with Spark/PySpark is a plus.
- Experience with scikit-learn, XGBoost/LightGBM; basic understanding of MLflow; containerization basics.
- Good grasp of statistics, A/B testing, and feature importance; curiosity about model risk and governance.
- Strong documentation and collaboration habits.
