Strategy, and the orgs that ship it.
Innovative engineering executive with 17+ years of experience (6+ leading engineering organizations, 5+ managing engineering managers), currently leading Amazon's Intent Personalization org of 100+ engineers and applied scientists. My teams deliver real-time, large-scale personalized recommendations powering customer-facing Search, Homepage, Cart, and Detail Page experiences for 500M+ customers globally, contributing $12B+ in ARR in 2025 and accelerating Amazon's topline growth.
I define and drive the multi-year technical vision for customer understanding and personalized recommendation engine at scale, leveraging ANN-based retrieval, transformer-based ranker and NLP to deliver nuanced product and intent understanding. My influence extends across 200+ engineers, scientists, and stakeholders in Search, Ads, and Home, aligning cross-org investments to scale innovation in personalization and advertising.
I excel in building high-performance AI organizations that combine cutting-edge innovation with operational excellence. I am passionate about mentoring engineers and managers, fostering collaborative environments where transformative AI solutions can flourish and drive significant business impact.
Where I've led.
- Lead a 100+ person Personalization Intent organization of Engineering Managers, Principal Engineers, and Applied Scientists, setting strategy, scaling the organization, and delivering AI-powered shopping experiences across Amazon.
- Define the multi-year vision and roadmap for Amazon's recommendation and customer intent platforms, contributing over $12B in annual attributed revenue (2025, +120% YoY).
- Own Amazon's real-time customer intent platform (500K RPS / 45B daily transactions) powering personalization across Search, Homepage, Cart, and Detail Pages.
- Lead strategy, development, and adoption of an AI agent platform for autonomous engineering and operational workflows: persistent memory, multi-agent orchestration, agents that monitor systems, triage issues, and support investigations, and secure integrations across Slack, Quip, Microsoft 365, and AWS tools. Scaled adoption across Personalization and to multiple organizations.
- Led development of LLM-generated customer profiles that fused deep learning behavioral signals with semantic customer understanding to power last-mile ranking.
- Lead Amazon's next-generation AI recommendation platform, leveraging customer intent, embeddings, retrieval, and GenAI to deliver personalized recommendations and explainable shopping guidance across major customer experiences.
- Built an NLP/LLM-powered product intelligence platform that standardizes product understanding and catalog attributes, enabling Search, Rufus, and p13n systems at scale.
- Established a new strategic organization developing ML models and distributed infrastructure for Amazon's prompt-based conversational shopping experiences.
- Drive multi-year investment planning, organizational strategy, and operational planning, partnering with senior leadership on long-term priorities.
- Promoted from Principal Engineer to Engineering Manager in Q1 2020; led 15 global engineers building a recommender engine on Google Vertex AI with custom embeddings.
- Built event-driven microservices and operated Kubernetes (K8s) clusters for high availability and failover; optimized pod scheduling and resource allocation, cutting downtime 40%.
- Implemented gRPC microservices for low-latency communication, cutting request latency 50%.
- Developed distributed pricing engines, personetics engine, deposit engine and real-time alerts for the Digital Wealth team.
- Led backend development of an AI-powered trust and safety web filtering system serving over 10 million students globally.
- Introduced real-time content classification using NLP and ensemble-based URL scoring, cutting review delays by 70%.
What I've built.
Help Me Decide (HMD)
Single-threaded owner of Amazon's GenAI shopping decision feature. $451M OPS in a three-week A/B test — the #1 Amazon weblab of 2025, cited in Q3 earnings.
Read the deep dive →GenRecs — Personalized Recommendations
Amazon's GenAI recommendation primitive turning natural-language intent into ranked products with a written rationale each. ANN retrieval over 100M+ items, serving 200M+ customers.
Customer360 (C360)
LLM-generated natural-language memory layer of Amazon's central customer memory system — standardizing data across Stores, Alexa, Prime Video, Music, Books, Audible, and Ads into structured data, embeddings, and memories any team or AI agent can query.
Contextual Messaging Service (CMS)
GenAI primitive that explains any recommendation across Amazon (Homepage, Search, Detail Page) using only observed customer signals. Adopted by multiple Store teams.