Senior Machine Learning Engineer (People Analytics / HR Analytics)
Type: Full-time
Type: Full-time
Location- Remote
Experience- 65+ years
About the role
We are hiring a Senior Machine Learning Engineer specialised in people analytics to deliver a privacy-first workforce wellbeing prediction capability within a large-scale analytics programme. You will build an ML model on fully anonymised HR and operational data that surfaces team-level wellbeing and burnout risk indicators, with explainability and fairness built in from day one. Privacy is not a constraint on this role; it is the defining design principle.
What you'll do
- Engineer features from anonymised workforce data: attendance patterns, leave utilisation, overtime concentration, shift rotation intensity, and engagement signals
- Develop, validate, and deploy a burnout risk prediction model on Databricks against a defined F1 acceptance target, producing team-level and organisational-unit-level outputs only
- Implement SHAP explainability attributing risk to anonymised feature groups without exposing any individual-level data
- Build bias detection and fairness checks across workforce groupings, in collaboration with a dedicated Data Privacy Lead
- Design alerting logic that triggers team-level notifications at configurable thresholds
- Implement MLOps: MLflow tracking, CI/CD deployment, drift monitoring, documented retraining
- Support privacy-focused SIT gates, UAT, documentation, and knowledge transfer
What you'll need
- 5+ years of ML engineering experience, including HR/people analytics or other sensitive-data domains
- Demonstrated experience with privacy-preserving ML: anonymisation, aggregation-only outputs, k-anonymity concepts
- Hands-on SHAP (or equivalent) explainability experience in production models
- Strong Python; production Databricks and MLflow experience
- Sound judgment on the ethics of workforce analytics and the ability to defend design choices to HR, security, and legal stakeholders
Nice to have
- Databricks certification (certification support provided)
- Experience with imbalanced classification and rare-event prediction (F1-driven evaluation)
- Experience working alongside privacy officers or under formal privacy impact assessments
- Prior delivery experience in the GCC region
Interested candidates share their resume at [Confidential Information]
