Sarah – NumPy, Pandas, Tensorflow

An internal market intelligence platform for the automotive aftermarket industry that combined sales, vehicle, demographic, and geographic data to estimate localized market potential, generate demand and segmentation insights, and support data-driven market expansion and commercial prioritization decisions.

• Built and refined the analytical logic behind market sizing and prioritization;
• Designed clustering and segmentation approaches for store and region classification;
• Developed demand estimation methodologies at multiple hierarchy levels and translated them into stakeholder-facing outputs;
• Helped create KPI frameworks and scalable analytical workflows that supported expansion and prioritization decisions.

A predictive customer potential and store segmentation platform that estimated customers’ untapped spending capacity, classified store performance across multiple business models, and enabled sales teams to prioritize high-value growth opportunities.

• Designed and implemented a Z-score-based bell curve methodology to classify store performance across the distribution network;
• Developed a robust Python pipeline to consolidate fragmented driver activity and sales data across multiple fiscal years;
• Engineered features reflecting customer behavior and regional market density to improve model accuracy;
• Automated the data aggregation process from nested directory structures, reducing manual data preparation time by over 15 hours per month.

An internal inventory alerting platform that monitored daily store and warehouse stock levels, generated dynamic thresholds, and flagged overstock and understock risks to support replenishment and inventory optimization decisions.

• Defined the MVP approach and translated inventory monitoring needs into analytical rules;
• Built logic to detect low- and high-inventory conditions using historical patterns rather than fixed generic cutoffs;
• Structured the project so it could evolve from a quantity-only MVP into a richer forecasting and optimization solution;
• Helped convert raw store-level data into actionable operational signals.

An internal analytics product focused on measuring store-level business performance through financial and operational KPIs, enabling leadership and business stakeholders to track performance, compare stores, identify business drivers, and improve reporting accuracy by resolving longstanding data discrepancies between source systems and reporting layers.

• Resolved the discrepancies and delivered the final product in 2 months;
• Delivered stakeholder-facing analytics tied to store performance and business outcomes;
• Helped translate operational data into decision-ready reporting;
• Supported performance analysis with a structured KPI lens;
• Worked closely with business users to make outputs practical and actionable.