Demand forecasting, pricing, personalization, GenAI, and supply chain intelligence. Built and deployed at Fortune 100 retailers.
Inconsistent forecasts across merchandising, supply chain, and finance at a 2,000+ store retailer. Overstocks, stockouts, idle transport.
Unified forecasting for 200K SKUs across stores and online channels including omnichannel inventory positioning. Scaled team from 5 to 45.
$3-4B in free cash flow through inventory optimization.
Traditional category hierarchies miss context-dependent product relationships.
Learned product representations that capture substitutability and complementarity from transaction data, feeding into forecasting, pricing, and assortment.
Poorly located warehouses after acquisitions. Inventory arriving too early or too late.
Simulation-based network design combined with real-time inventory control across the full multi-tier network.
Inaccurate inventory counts causing phantom stockouts and lost sales.
Continuous inventory estimation by triangulating multiple data sources without requiring manual counts.
Single-source procurement exposed to supply chain disruptions.
Continuous supplier monitoring with automated risk playbooks that trigger contingency actions.
Multiple prior attempts at pricing optimization had failed. Merchants lost trust.
Embedded optimization directly in merchant workflows. Merchants see recommendations alongside projected margin impact.
$100M+ profit impact.
True incremental lift from promotions was impossible to isolate. Supply chain had no visibility into upcoming promotions.
Promotional lift models integrated with demand forecasting so promotional plans automatically adjust supply chain planning.
A major grocer needed analytics to scale its advertising business.
Experimentation engines, ad inventory forecasting, and budget-to-segment optimization.
30% CAGR at 50% margin.
Generic campaigns sending the same offer to every customer.
Personalization engines determining the right product, offer, and message for each customer at each moment.
Customers churn silently. By the time they are "lost," it is too late.
Unified customer view with next-best-action engine recommending the optimal intervention per customer.
ROAS attributes sales to a single touchpoint, missing the full customer journey.
Attribution model that decomposes the contribution of each channel and touchpoint across the full customer journey.
Customers cannot describe visual attributes in keywords.
Computer vision that matches uploaded images to the product catalog, ignoring background distractions.
Multiple teams experimenting with GenAI independently without governance.
25-person CoE owning the governed platform and guardrails. Production use cases across marketing, product content, search, call center, and store operations.
Six production GenAI applications: marketing copy generation, product image creation, PDP content for 200K+ SKUs, semantic search, call center agent augmentation, and an internal chatbot for store and warehouse employees.