Sai Prasad

About Sai Prasad

I build products/processes that simplify complex and ambiguous workflows. My experience spans product strategy, AI automation, forecasting, and operations, with a focus on turning scattered processes into clear and scalable systems. I work at the intersection of data, decisions, and user needs, connecting disconnected sources, improving forecast accuracy, and automating repetitive analysis to create more consistent and reliable insights. I collaborate closely with engineering, analytics, operations, and business teams to define problems, build practical solutions, and drive alignment. I am especially interested in AI systems, data products, forecasting tools, and operational intelligence, and I enjoy working on problems where thoughtful product design and cross-functional execution create real impact.If you’re building AI or data-driven products and want fewer spreadsheets named “FINAL_v7_really_FINAL,” let’s connect.

Experience

1. Demand Planner Product

   Amazon

   Dec 2023 — Present · Seattle, WA, US

   Built an automated instock intelligence system unifying 12 Fresh/WFM dashboards into a single top-50 ASIN tracker, integrating inventory, forecasts, confirmation rates, and risks across Stores/FCs. Eliminated manual reporting and standardized insights through an AI narrative engine.• Owned demand planning for Amazon’s Deal of the Day program, developing category and placement-based demand models for 15+ high-elasticity items. Supported promotions that generated ~10.6K new customers, 28K+ orders, and 10.5M+ impressions while improving accuracy using category-specific elasticity.• Designed and deployed a multi-vertical calibration engine automating trend detection and bias correction across Fresh. Reduced p90 underbias 17%→13%, overbias 20%→16%, tightened forecast spreads 7→2, and surfaced zero-forecast blind-spot items.• Led forecasting and instock readiness for Prime Big Deal Days, modeling promo and coupon lifts (26% online,~100% in-store) and supporting events that drove ~62% of in-store PD volume. Rebuilt lift assumptions using multi-year data and aligned pricing, retail, instock, and marketing inputs to deliver accurate forecasts under tight timelines.• Owned ASIN and basket-level demand modeling for Prime Big Day Deals across 303 SKUs, informing a 13.6M-unit forecast (+22.5% vs plan). Built lift models (500–900% for free-item offers), enabling upstream buying and inventory positioning.• Led produce promo forecasting for 10 months, delivering the highest instock performance in Fresh (91% across Stores/FCs) while managing seasonal and promotional volatility.

Education

• Visvesvaraya Technological University

  Bachelor of Engineering - BE, Mechanical Engineering

  2010 — 2014

• Babson College

  Master of Business Administration - MBA

  2018 — 2020