Ashwin Rajagopalan

About Ashwin Rajagopalan

Senior Data Engineer with 12+ years of experience designing, building,and owning large-scale, cloud-native data platforms.Currently at Amazon, I work on high-volume analytics and event-drivensystems processing $100B+ in spend data. My core expertise lies inSpark-based data pipelines, lakehouse architectures (Hudi, Iceberg),real-time ingestion, and low-latency data systems using DynamoDB andOpenSearch.I have extensive experience enabling machine learning and GenAIcapabilities on top of data platforms, including ML-driven taxonomyclassification, large-scale deduplication using text and vector search,and embedding-based enrichment pipelines. I’ve also integrated LLM-basedautomation (via Bedrock) into production systems to improve data qualityand developer productivity.My work emphasizes end-to-end platform ownership, scalability,reliability, and data quality, with hands-on involvement acrossarchitecture design, performance optimization, and cross-team delivery.I enjoy solving complex data problems at scale and building robust dataplatforms that power critical products and business decisions.

Experience

1. Data Engineer 2

   Amazon

   Sep 2022 — Present · Hyderabad, IN

   Led development of Lantern Analytics Platform, an enterprise analytics system processing $100B+ spend data, reducing manual validation by ~80% and enabling ML-driven taxonomy classification.Architected Docket, a serverless, event-driven tax invoice ingestion platform handling 25K+ notifications/hour across multiple countries with 100% data accuracy and zero production incidents; used DynamoDB for event state tracking, retries, and recovery.Built backend services for Braavos, a self-service analytics platform, enabling autonomous report generation via dynamic query generation and orchestration.Designed and developed Supplier Genie, an AI-driven supplier data platform; architected OpenSearch text + vector indexes for semantic search and advanced deduplication.Optimized ETL pipelines to ingest ~290M records with embedding generation into OpenSearch in ~45 minutes, evaluating multiple embedding models for performance and quality.Integrated Bedrock-based LLM solutions for supplier enrichment and classification; built an AI-powered TestCase Generator to auto-generate BRDs and test cases.Developed reusable AWS CDK infrastructure constructs, reducing data pipeline setup from ~4 weeks to 3 days.Technical owner for cross-team initiatives influencing data platform architecture, search design, and GenAI adoption.

Education

• Jawaharlal Nehru Technological University

  Bachelor of Technology (B.Tech.)

  2009 — 2013

• International Institute of Information Technology Bangalore

  PG Diploma in Machine Learning & AI November 2019 - Course 4

  2019 — 2020

• International Institute of Information Technology Bangalore

  PG Diploma in Machine Learning & AI November 2019 - Course 1

• International Institute of Information Technology Bangalore

  Machine Learning and Artificial Intelligence

• Liverpool John Moores University

  Master of Science - MS

Skills

• Matlab
• C++
• C
• Programming
• Windows
• Microsoft Excel
• Powerpoint
• Electrical Engineering
• Microsoft Office
• Engineering
• Teamwork
• Microsoft Word
• Electronics