Software Engineer | 4 years | Distributed System, Backend Development, Spring Boot, React, AWS

I am a Software Development Engineer (SDE II) with around 4 years of experience architecting high-throughput distributed systems, event-driven pipelines, and latency-critical access control platforms. My engineering background is rooted in financial and payment data infrastructure across tier-1 organizations including PayU (Wibmo) and Bank of New York (BNY). Throughout my career, I have specialized in building fault-tolerant microservices, enforcing zero-duplicate transaction processing, designing OAuth/Passkey security layers, and optimizing large-scale data queries under strict financial SLAs.

At PayU, I own the request/response authentication flows (OAuth, JWT, OTP, Out-of-Band, and Passkey/biometric authentication) for major payment networks, including Visa and Mastercard. These services are built for high availability and fault tolerance, replicated across 8 instances per microservice across dual data centers, load-balanced via Nginx, and fronted by CDNs such as Cloudflare and Akamai. To modernize credential management, I designed and delivered full-lifecycle Passkey registration and deregistration, expanding frictionless authentication coverage. Additionally, I eliminated redundant lookups and lowered CPU utilization on latency-critical authentication-routing paths by introducing Redis caching for bank-specific BIN configuration data. I also developed React.js admin dashboards and monitoring panels to provide clear operational visibility into challenge/response flows and cache performance.

Prior to PayU, I spent three years at Bank of New York (BNY) leading core backend initiatives. I designed an end-to-end Kafka-based financial data ingestion platform spanning 6 microservices that processes 40,000+ daily transactions with idempotency, automated retries, and dead-letter queues (DLQ) to guarantee zero duplicate payments under strict financial SLAs. I engineered and scaled MongoDB and Elasticsearch services managing 500GB+ of financial data across 40+ collections, utilizing compound indexing, aggregation pipelines, versioning, and change streams for real-time data consistency. To accelerate downstream reconciliation workflows, I built batch ingestion pipelines using Apache Spark for clients such as BlackRock. In performance engineering, I designed a multithreaded investor entity-resolution engine utilizing weighted similarity matching algorithms, reducing KYC reconciliation runtimes from 30 minutes to 4 minutes (an 87% reduction).