Lead Data Scientist (AI/ML)
Role summary
We are seeking a Lead Data Scientist to spearhead our supply chain demand forecasting and root cause analysis platform. This is a senior, hands-on individual contributor role focused on implementing, validating, and maintaining the full ML pipeline. You will leverage expert-level Python skills, deep experience with ensemble methods and probabilistic forecasting, and strong statistical foundations. Proficiency in SQL and version control is essential. The role requires 9-12 years of data science/ML experience, with at least 3 years in time-series forecasting or supply chain analytics within a commercial setting. A Master's or PhD in a quantitative field is preferred, though equivalent experience will be considered.
We are hiring a Lead Data Scientist to be the primary technical engine of our supply chain demand forecasting and root cause analysis platform. This is a hands-on senior individual contributor role with significant ownership — you will implement, validate, and maintain the full ML pipeline, working closely with the US-based Senior Manager.
Required Qualifications
Experience
- 9–12 years of hands-on experience in data science or machine learning — with a strong emphasis on Python-based ML engineering in production environments
- 3+ years of experience with time-series forecasting or supply chain analytics in a commercial context
- Demonstrated experience building end-to-end ML pipelines from raw tabular data through model output and reporting — not just notebook prototyping
- Experience working in cross-functional teams with stakeholders across business, IT, and analytics; ideally in a consulting or professional services environment
- Track record of delivering high-quality, well-documented, reviewable code in a team setting
Technical Skills
- Expert-level Python: scikit-learn, pandas, numpy, scipy, joblib — able to write production-grade, optimised code for large datasets
- Deep hands-on experience with ensemble methods: gradient boosting (GBM, XGBoost, LightGBM) and Random Forest — including hyperparameter tuning and performance diagnostics
- Proficiency in quantile regression and probabilistic forecasting: building tree-level percentile prediction intervals, measuring PI coverage (Winkler score, pinball loss), and detecting quantile crossing violations
- Strong statistical skills: KS 2-sample tests, ACF/PACF analysis, change-point detection, IQR outlier detection, Pearson/Spearman correlation
- Proficiency with SQL for data extraction, transformation, and validation
- Familiarity with version control (Git), experiment reproducibility (SEED management, config-driven pipelines), and collaborative development workflows
Education
- Master's degree or PhD in Data Science, Statistics, Computer Science, Machine Learning, Operations Research, or a related quantitative field
- Bachelor's degree with equivalent industry experience in a quantitative discipline considered
Preferred Qualifications
- Experience with intermittent demand modelling: Croston method, SBA, ADI and CV² classification for routing parts to appropriate forecast models
- Experience with reconciliation frameworks: bottom-up and top-down forecast reconciliation, MinT reconciliation, hierarchical coherence
- Familiarity with MLflow, DVC, or equivalent tools for experiment tracking and pipeline orchestration
- Experience with cloud platforms (AWS SageMaker, Azure ML, or GCP Vertex AI) for scalable model training and deployment
- Knowledge of S&OP processes, IBP (Integrated Business Planning), and multi-echelon inventory theory
- Experience building user-facing analytical tools or dashboards — ideally with some exposure to full-stack data product development
- Contributions to open-source ML projects or published work in forecasting, supply chain analytics, or applied ML
