Jan 2023 – Present · 3 yrs 2 mos

• Leading the global delivery of inventory, size, and forecasting platforms for a top apparel brand across 80+ countries, while running high-bar POCs with a leading athleisure company. Navigating complex stakeholder maps and tough SLAs, flagging risks early, and landing pragmatic solutions drawn from prior large-scale rollouts.
• Products - Implementing InventorySmart, SizeSmart, and ForecastSmart end-to-end across 80+ countries; aligning assortments, regional calendars, size curves, and DC constraints—so weekly allocations stay steady and stores stay ready.
• Running parallel POCs with a major athleisure player—setting sharp success criteria (availability lift, size-curve accuracy, forecast error), designing control tests, and converting wins into scale playbooks.
• Partnering with VPs, CXOs, and Senior Directors across Supply Chain, Merchandising, Planning, IT, and Finance—chairing executive steercos, publishing RAID and value scorecards, and keeping delivery “no surprises.”
• Meeting demanding SLAs on uptime, response, and allocation cycle time—instrumenting live health, incident run-books, and clean on-call lanes.
• Surfacing upstream risks early—data drift, assortment churn, DC capacity, pack/size misalignment—and de-risking via phased waves, feature flags, and reversible launches.
• Strategy of Work -
• Building repeatable delivery: pilot → A/B vs control → phased scale-up, with rollback levers defined on day one.
• Hardening data contracts and environments so peak seasons stay calm.
• Coaching cross-functional teams to focus on the 5% of exceptions that drive 80% of outcomes.
• Early outcomes (ongoing)
• Lifting on-shelf availability for priority SKUs while lowering aged inventory exposure.
• Improving size-curve fit and reducing size-based stockouts in trial markets.
• Cutting manual overrides and planner fire drills during seasonal resets.

Jan 2021 – Dec 2022 · 1 yr 11 mos

• I led end-to-end rollouts of an enterprise inventory planning platform across three retailers:
• a global multi-brand luxury house,
• the largest specialty jeweler, and
• a North American value retailer (30k+ stores).
• Luxury multi-brand: unified ERPs and product/location hierarchies, encoded brand presentation rules, tuned allocation logic for capsule drops and boutiques, and moved planners to exception-first workflows. Result: higher on-shelf for hero SKUs and fewer fire drills at launches.
• Specialty jeweler (sparse demand, high ticket sizes): built attribute-pooled forecasting, lost-sales estimation, and service-level targets tied to ticket size and lead time. Result: fewer stockouts on winners and a cleaner exit path for slow movers without starving stores.
• Value retailer at national scale: operationalized weekly wave planning across 30k+ stores with DC and truck constraints, carton rounding, and store-capacity guardrails. Result: stable high-volume allocations and smoother seasonal resets.
• Senior-stakeholder engagement —
• Ran executive steering with multiple VPs and CXOs (Supply Chain, Merchandising, Technology, Finance) and Senior Directors.
• Presented quarterly value reviews, secured scope decisions, and delivered predictable outcomes with crisp logs and transparent KPIs.
• Delivery —
• Translated boardroom priorities into planner-friendly rules: service levels, presentation minimums, launch throttles, DC/store capacity, and override guardrails.
• Drove adoption via pilot → A/B vs control → phased scale-up with rollback levers.
• Instituted exception playbooks so teams focused on the 5% of decisions that move 80% of results.
• Stood up runbooks, SLAs, and change control to keep peak seasons error-free.
• Selected outcomes —
• Lifted availability on priority SKUs while reducing aged-inventory exposure.
• Cut manual overrides and firefighting during seasonal peaks.
• Freed planner time through exception-driven workflows; raised confidence in weekly allocations.

Jun 2019 – Dec 2020 · 1 yr 6 mos

• ✓Worked end to end on building a web-based buy optimization product BuySmart for a US fashion retailer. The product optimizes the buying of products from the vendor, at a set expenditure and it's distribution across specific stores, maximizing revenue. The product BuySmart uses the MILP model for optimizing the purchase of goods and store allocation and CNN for image mapping.
• ✓Promotion and price optimization for the largest US music retailer. The optimization includes several analysis pieces - price elasticity, affinity analysis and cannibalization.
• ✓Markdown analysis for a US music retailer and a fortune 500 US Fashion retailer forecasting the discount points and inventory on hand the retailers should have on each of their stores for clearance products.

Jun 2017 – May 2019 · 1 yr 11 mos

• ✓Worked end to end on building a web-based promotion optimization product PromoSmart, which decides the discount points and its scheme during various critical promotion campaigns like Black Friday, Christmas, President's Day etc maximizing margin. The tool decides the best marketing techniques to be used for the campaign. It has the capability to compare past promotions and deep dive into its trend. It helps the user create promotion campaigns and forecast the overall margin gain 6 months in advance. The forecasting model consists of an ensemble of various models like Linear regression, Poisson regression, XGBoost and Lasso-Ridge regression.
• ✓Worked on Fraud and Geo Analytics for the largest American Automobile Insurance company identifying various frauds and costs incurred due to each. Non-linear directionality reduction algorithm (t-sne & UMAP) for variable assortment and various predictive models (Arima, Facebook's Prophet & Multiple Regression) for forecasting the cost due to fraud for the future months.
• ✓Pricing analysis for the largest law firm in the US for its Real Estate cases and boosting bottom-line for the firm.