Piyush Tripathi

Architected scalable Java/Spring Boot back-end microservices for an enterprise inventory management platform and integrated external Offer APIs. Built a fault-tolerant batch pipeline on Azure Service Bus that processes millions of records daily, cutting processing time by ~90% (hours to minutes). Designed a multi-tenant Azure SQL/MS SQL Server database with isolation and query optimization to improve read/write throughput. Implemented Auth0/OIDC, Spring Security, and JWT role-based access control; delivered scheduled jobs to synchronize inventory, pricing, and marketplace data in near real time. Applied circuit breaker, retry, and idempotency patterns for resilience in service-to-service and message-queue communication. Delivered enterprise IoT back-end systems (REST Gateway + Authorization APIs) deployed on Azure Kubernetes Service (AKS) for real-time processing. Led CI/CD automation plus security scanning and centralized monitoring/logging to reduce MTTR during incidents.

Architected scalable Java/Spring Boot backend microservices for an enterprise inventory management platform (Goodwill Inventory Management System), integrating Amazon product and marketplace offer APIs. Built a high-volume fault-tolerant batch pipeline on Azure Service Bus that processed millions of records daily, reducing batch processing time by ~90% (hours to minutes). Designed a multi-tenant database architecture on Azure SQL/MS SQL Server with isolation across client tenants, applying clean architecture and query optimization for improved read/write throughput. Implemented Auth0 OIDC with Spring Security and JWT-based role access control, and built scheduled jobs to synchronize inventory, pricing, and marketplace placement. Applied circuit breakers, retry, and idempotency patterns for resilient service-to-service and message-queue communication. Delivered enterprise IoT backends (Honeywell Forge Connect) using REST gateways and authorization APIs, deploying fault-tolerant microser

Designed and led development of a batch payment transaction processing and reconciliation system for Capital One Bank, including an orchestration service coordinating multiple controllers with zero payment failures post-development. Migrated legacy services to Spring Boot 3.x/Java 17 and built cloud-native, event-driven components on AWS Lambda and ECS, cutting transaction processing time from hours to minutes using PostgreSQL/Aurora. Built distributed event-driven systems on AWS using Kafka-based architecture and Java multithreading to handle thousands of simultaneous batch payment transactions with zero-failure performance. Optimized high-volume record processing into Cassandra (hours to minutes), implemented Redis caching and Drools rules engine to reduce API response times and enable business-configurable transaction logic. Reviewed code and mentored junior engineers in an Agile/Scrum team, increasing unit test coverage and shortening defect turnaround across payment services.