Novulis Global Solutions - Data Scientist

Job description
Job description:Job Description:We are seeking a highly skilled Data Scientist to join an innovative team that successfully developed a 05 month baseline sales forecasting solution during the Personal Health Innovate2Win Hackathon. Building on this success, we are now extending the solution across additional accounts to evaluate the feasibility of creating highly reliable and precise forecasting models using enriched data from the direct commercial channel. This role offers the opportunity to work at the intersection of advanced analytics, machine learning, commercial planning, and business strategy. The ideal candidate will quickly understand existing methodologies, enhance current forecasting capabilities, and contribute to the next generation of forecasting solutions both in terms of model performance and feature innovation. While this opportunity begins as a gig assignment, a successful outcome could lead to the establishment of a long-term internal capability within the Personal Health organization.Key Responsibilities:1. Forecasting Model Development:- Design and implement bottom-up demand forecasting models with forecasting horizons of up to 05 months.- Work across SKU-level and hierarchical forecasting structures (SKU Article Group), addressing challenges such as intermittent demand and sparse sales histories.2. Data Understanding & Feature Engineering:Analyze, integrate, and derive insights from diverse commercial data sources, including:- Sell-out / POS data (daily and weekly)- Sell-in data- Stock-on-hand and weeks-of-stock metrics- Promotion and event calendars- Product lifecycle information (Phase-In / Phase-Out, NPI/PIPO)- Planning inputs from Excel-based planning tools and BI dashboardsAdditionally, you will:- Explicitly address data limitations such as weak promotional correlations, visibility uncertainty, and marketplace effects (e.g., Buy Box dynamics and retail media impact).- Assess market insights and incorporate relevant signals into feature engineering and model development.3. Evaluation, KPIs & Validation:- Measure model performance using the same KPIs applied in manual planning processes, enabling fair side-by-side comparisons.- Evaluate metrics such as forecast accuracy and forecast bias.- Validate models to ensure performance meets or exceeds planner-driven approaches before broader adoption.- Prevent overfitting through careful treatment of soft variables and uncertain demand drivers.4. Explainability & Decision Support:- Implement explainable AI techniques such as SHAP and LIME to ensure model outputs are transparent and interpretable for business stakeholders.Support planner-controlled what-if simulations, including:1. Uplift assumptions2. Promotional strength scenarios3. Inventory guardrails- Contribute to the development of LLM-based natural language explanations that help explain forecasts rather than generate them.5. Business Collaboration & Stakeholder Engagement:- Partner closely with commercial planners, demand planners, and insights managers to understand planning processes, operational realities, and business constraints.- Translate complex analytical outputs into actionable, decision-oriented business insights.- Participate in workshops, model reviews, and feasibility assessments with cross-functional stakeholders.Important Qualifications:Required Experience & Expertise:- Strong academic background in Data Science, Applied Mathematics, Statistics, or a related quantitative discipline.- Hands-on experience developing forecasting models for time series data, preferably within supply chain, demand planning, commercial, or retail environments.- Proficiency in Python (or a similar programming language) for data analysis, machine learning, and model development.- Experience working with sparse, noisy, and intermittent demand datasets.- Strong understanding of forecasting validation methodologies, bias measurement, and accuracy metrics.- Familiarity with Explainable AI techniques, including SHAP and model interpretability frameworks.- Experience with hierarchical forecasting across multiple levels such as SKU, product group, and customer hierarchies.- Demonstrated ability to collaborate effectively with both technical and non-technical stakeholders.Role: Data ScientistIndustry Type: IT Services & ConsultingDepartment: Data Science & AnalyticsEmployment Type: Full Time, Temporary/ContractualRole Category: Data Science & Machine LearningPreferred Key Skills:- Time Series, Analysis, Machine Learning, Python- Retail, LSTM, ARIMA, Demand Forecasting, Supply Chain, Forecasting