Mahmudur Rahman
Applied Scientist @Amazon || PhD @UMBC
- Role
- Applied Scientist at Amazon
- Location
- Seattle, WA, US
- LinkedIn followers
- 500 followers
About Mahmudur Rahman
I am an Applied Scientist in the Central Seller Fulfillment team at Amazon, where I drive…
Experience
Applied Scientist
Sep 2024 — Present · Seattle, WA, US
Delivered causal inference-based personalized recommendation system serving 4M+ sellers across20+ global marketplaces, directly improving seller experience and business outcomes- Owned end-to-end productionization of ML causal recommendation model, replacing legacy systemand scaling from 4 to 20+ marketplaces with measurable performance improvements- Built and deployed production-grade ML pipelines using AWS Step Functions, Cradle, EMR, and SageMaker, ensuring high availability and fault tolerance for critical seller-facing systems- Designed scalable data infrastructure processing petabyte-scale datasets, integrating multiple AWSservices (Lambda, EMR, SageMaker) to support weekly inference and batch training workflows- Designed customer-obsessed explainability framework providing actionable, business-aligned insights that enable sellers to optimize interventions and drive growth- Invented flexible, reusable model framework adopted by multiple Amazon teams, demonstrating scalability and cross-team collaboration while maintaining high engineering standards
Education
University of Maryland Baltimore County
Doctor of Philosophy - PhD
2019 — 2024
University of Dhaka
Bachelor of Science (BSc)
2010 — 2014
University of Maryland Baltimore County
Master of Science - MS
2019 — 2023
University of Dhaka
Master of Science (MSc)
2015 — 2016
Find verified contacts for anyone on LinkedIn
Unifers gives sales teams verified emails and direct dials, enriched profiles, and outreach that lands in the inbox.
Free plan included · No credit card required
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.