The Data Science Lead is a pivotal technical leader responsible for architecting and deploying production-grade Machine Learning (ML) and Artificial Intelligence (AI) solutions. You will steer the development of sophisticated predictive models within a robust CI/CD/CT (Continuous Training) framework. Leveraging the Azure ecosystem (Databricks, Spark, Azure ML), you will transform big raw data into scalable intelligence, ensuring that models are not just "notebook experiments" but resilient enterprise assets.

Responsibilities

Architectural Leadership: Lead the design and delivery of end-to-end ML systems, prioritizing MLOps principles to ensure model reproducibility, auditability, and scalability.

Full-Lifecycle Development: Oversee the journey from hypothesis and Exploratory Data Analysis (EDA) to feature engineering, model selection, and production deployment.

Cross-Functional Synergy: Act as the technical bridge between Data Engineers (for ETL/Feature Store optimization) and Business Stakeholders (to translate KPIs into objective functions).

Infrastructure Automation: Architect automated pipelines for data validation, model profiling, and hyperparameter tuning using Azure Machine Learning Services.

Governance & Monitoring: Establish rigorous monitoring for Data Drift and Concept Drift, ensuring model performance remains optimal post-deployment.

Requirements

Experience: 5–8 years of total experience, with 4+ years specifically in a hands- on Data Science role and 2+ years leading technical teams or complex projects.

Education: BE/BS or MS/PhD in Computer Science, Statistics, Mathematics, Physics, or a related quantitative field.

1. Advanced Modeling & Mathematics- Deep Learning & Classical ML: Proficiency in supervised/unsupervised learning, including Gradient Boosted Trees (XGBoost/LightGBM), Random Forests, and Neural Networks.

Statistical Rigor: Mastery of hypothesis testing, Bayesian inference, and error analysis. Ability to design complex experiments and A/B tests.

Time Series & Forecasting: Experience with advanced forecasting (Prophet, ARIMA, or LSTM networks) is highly desirable for Supply Chain/Revenue applications.

Optimization: Knowledge of loss function customization and optimization algorithms (Gradient Descent, Genetic Algorithms).

2. Engineering & MLOps (The 'Lead Edge)- The Stack: Advanced proficiency in Python (Pandas, Scikit-learn, PySpark/TensorFlow) and SQL.

Big Data: Hands-on experience with PySpark and Databricks for distributed processing of petabyte-scale datasets.

Orchestration: Expert-level knowledge of MLFlow for experiment tracking and Kubeflow or Azure Pipelines for orchestration.

Deployment: Experience with containerization (Docker/Kubernetes) and deploying models batch inference jobs.

Nice to Have

Strategic Translation: The ability to take an ambiguous business problem (e.g., 'We are losing margin in the Midwest') and translate it into a specific ML problem (e.g., 'A multi-classification churn model with a SHAP-based interpretability layer').

Technical Mentorship: A proven track record of conducting code reviews, promoting best practices in ''Clean Code', and upskilling junior data scientists.

Agile Advocacy: Deep familiarity with Agile/Scrum methodologies, specifically how to adapt 'Sprint'; cycles to the non-linear nature of Research & Development.

Influence & Stakeholder Management: The 'soft power' to explain complex model trade-offs (e.g., Precision vs. Recall) to non-technical executives to drive data-driven decision-making.