Machine Learning Engineer – Risk Scoring, Model Monitoring
Posted 1hrs ago
Employment Information
Report this job
Job expired or something wrong with this job?
Job Description
Machine Learning Engineer building explainable risk-scoring and monitoring models for Chiri’s healthcare and pharmacy-finance platform. Developing reliable gradient-boosted pipelines with drift detection and controlled deployment.
Responsibilities:
- Develop and evaluate predictive models using XGBoost, LightGBM, and CatBoost for risk scoring, classification, and related decision support
- Compare machine learning performance against deterministic rules and historical benchmarks
- Detect changes in input distributions, prediction behavior, and model accuracy, with alerts when performance falls below defined thresholds
- Build feedback loops using newly available outcome data, with retraining and validation procedures
- Identify and prevent target leakage, temporal leakage, proxy leakage, and unintended bias in model features
- Implement reproducible model pipelines, monitoring, performance reporting, explainability, and controlled model deployment
- Determine where machine learning adds measurable predictive value to a healthcare and pharmacy-finance platform
- Explain model performance, limitations, and business implications to technical and nontechnical stakeholders
Requirements:
- Strong production machine learning experience using Python
- Hands-on expertise with XGBoost, LightGBM, CatBoost, or comparable gradient-boosted tree frameworks
- Deep understanding of statistics, feature engineering, model evaluation, calibration, and temporal validation
- Experience detecting and managing data drift, concept drift, model degradation, and retraining triggers
- Proven experience preventing data leakage and identifying proxy variables that compromise model validity
- Experience developing production ML pipelines with automated monitoring, testing, and reproducibility
- Ability to explain model performance, limitations, and business implications to technical and nontechnical stakeholders
- Experience with healthcare, pharmacy, financial risk, credit scoring, or claims analytics
- Experience with delayed outcome labels and historical transaction datasets
- Experience deploying interpretable models in regulated environments
- Experience combining deterministic rules with probabilistic models
Benefits:
- The opportunity to help establish a rigorous, measurable approach to machine learning in a high-stakes business environment.













