At Johnson & Johnson, I work on building scalable backend services for an AI-driven clinical knowledge retrieval platform designed to help researchers and clinicians quickly access insights from large collections of medical literature and clinical trial data. I develop microservices using Java, Spring Boot, and reactive frameworks, design secure REST and GraphQL APIs, and integrate AI-powered document retrieval workflows using LangChain, Hugging Face models, and vector databases to enable semantic search across millions of documents. My work also involves improving system reliability and observability using tools like Prometheus, Grafana, and OpenTelemetry, while ensuring secure access control, high availability, and performance across distributed cloud-native systems.

• Programmed backend microservices using Java (Spring Boot) for digital wallet top ups, merchant payments, and transaction history APIs, designing

scalable, fault-tolerant services with sharding, partitioning, and concurrency controls, supporting 98% SLA, sub-250ms API response times.

• Designed and implemented secure RESTful and GraphQL APIs with JWT/OAuth2 authentication and PCI DSS-compliant tokenization for card and

wallet data, incorporating efficient data structures and algorithm optimizations to reduce latency and sensitive data exposure risk by 98%.

• Formed responsive mobile front-end modules using React Native and TypeScript for wallet UI, payment flows, QR scanning, and real-time balance

updates with WebSockets and Redux, applying reactive programming and optimized state, increasing transaction completion rate by 22%.

• Engineered real-time payment pipelines with Apache Kafka, Redis caching for session & balance, and PostgreSQL for transactional consistency,

using queue partitioning, event-driven design, and distributed caching to achieve reliable sub-200ms balance updates during peak operations.

• Deployed cloud-native infrastructure on AWS using EKS (Kubernetes), API Gateway, Lambda serverless functions, and RDS Aurora with multi-AZ

failover, applying system design principles for auto-scaling, high availability, and disaster recovery, achieving 99.99% uptime.

• Established CI/CD pipelines with GitHub Actions, Terraform IaC, automated unit/integration tests (JUnit, Jest), and blue/green releases on

Kubernetes, integrating test-driven design to reduce rollback incidents by 60% and ensure production stability.

• Integrated third-party carrier APIs, ERP systems, and payment gateways with backend services to enable real-time shipment tracking, order

fulfillment updates, and transaction validation, improving on-time delivery visibility by 28%.

• Configured backend microservices using Node.js and Express.js for inventory management, order processing, and logistics workflows, reducing

order processing latency by 33% and supporting concurrent warehouse operations.

• Optimized MongoDB data models for warehouses, shipments, and order transactions, implementing indexes, aggregation pipelines, and sharding

strategies, reducing inventory query times by 42%.

• Formulated interactive React.js dashboards for supply chain managers to monitor shipments, track inventory levels, and generate analytical

reports, increasing operational decision-making efficiency by 22%.

• Led migration of legacy order and inventory data to MongoDB Atlas, designing ETL pipelines that ensured zero data loss, improved multi-region

replication, and reduced reporting latency by 30%.

• Orchestrated CI/CD pipelines using Jenkins and GitHub Actions, containerized Node.js services with Docker, and deployed on Kubernetes with

Helm charts, enabling automated and consistent deployments across staging and production.

• Executed automated unit and integration tests using Jest and Supertest, achieving 80%+ backend coverage while enforcing JWT-based role-based

access control for customers, admins, and vendors, reducing production defects by 33%.