I'm a backend-focused software engineer specializing in enterprise Java, Spring Boot, and distributed systems. I build high-throughput APIs, event-driven services, and automation solutions designed for maintainability, resilience, and secure operation in production.

My work spans application modernization, database upgrades, asynchronous processing, CI/CD automation, observability, performance engineering, and production-grade AI-assisted workflows. I care about systems that are easy to operate—not just easy to demo—and services that do not wake anyone up at 3 AM.

• 🏗️ Build scalable, maintainable services using clean architecture and proven design patterns
• ⚡ Design event-driven and asynchronous flows with Kafka and message queues
• 🔄 Automate manual business processes and software delivery workflows
• 🔐 Treat security, validation, and sensitive-data protection as design requirements
• 📊 Use logging, health checks, tests, and load analysis to improve production reliability

Microservices · Clean Architecture · SOLID · Design Patterns · API Security

Event Streaming · Partition Strategies · Exactly-Once Semantics · Caching · Session Management · Rate Limiting

Pipeline as Code · Multi-Environment Delivery · Rollback Strategies · Structured Logging · Health Checks · Unit & Integration Testing

AI Agents · Production Prompt Engineering · Backend-Heavy Full-Stack Delivery

A production-style commerce ecosystem spanning a customer storefront and an enterprise order-management platform, connected through bidirectional Kafka event flows.

QuickBasket Storefront
        │  ORDER_PLACED · oms.orders.inbound
        ▼
    Apache Kafka
        │
        ▼
Enterprise OMS ── Order · Inventory · Payment · Shipping · Analytics
        │  ORDER_STATUS · oms.orders.status
        ▼
    Apache Kafka
        │
        ▼
QuickBasket ── Audit History · User Notification · Live SSE Update

• QuickBasket E-Commerce System — Java 17, Spring Boot 3.2, React 19, MySQL, Redis, JWT security, Razorpay payments, Kafka, and server-sent events. Implements catalog, cart, checkout, orders, returns, notifications, and administration across a seeded catalog of 170+ products and 20 categories.
• Enterprise Order Management System — Java 17 and Spring Boot 3.5 modular monolith covering order lifecycle, inventory reservation, payment, shipping, analytics, JWT/RBAC, Flyway migrations, and a Kafka-backed transactional outbox.
• Publishes committed orders to oms.orders.inbound and returns lifecycle updates through oms.orders.status, with messages keyed by order number to preserve per-order partition ordering.
• Designs for at-least-once delivery through idempotent order intake, producer idempotence, auditable status history, and explicit failure-handling trade-offs.

View Storefront Repository · View OMS Repository · Kafka Integration Contract

A human-in-the-loop support workflow that combines enterprise Java engineering with grounded generative AI.

• AI Customer Support Assistant — Java 21, Spring Boot 3.3, Spring AI, PostgreSQL, Docker Compose, Spring Security, and JWT/RBAC.
• Implements ticket analysis, priority classification, provider-agnostic LLM integration, controlled tool calling, and agent approval or rejection of generated responses.
• Builds a RAG pipeline with document chunking, vector search, grounding, source citations, and a guard that avoids calling the model when relevant context is unavailable.
• Adds production-style controls through AI audit logs, Prometheus metrics, health indicators, retry and circuit-breaker policies, structured logging, and unit/integration testing with JUnit, Mockito, and Testcontainers.

View Repository

_{These are independent portfolio systems designed to demonstrate architecture and engineering practices; they are not representations of vendor-owned production platforms.}

• Modernized enterprise Java applications from Java 8 to Java 21, addressing deprecated APIs, module-system considerations, library compatibility, and regression coverage while maintaining a zero-downtime rollout.
• Delivered a MongoDB 8.x migration, adapting schema validation and aggregation pipelines while revisiting index strategy for compatibility and performance.
• Engineered Jenkins CI/CD pipelines that automate builds, tests, multi-environment deployments, and rollback procedures for safer, repeatable releases.
• Improved distributed-service operability through centralized structured logging, ELK-based analysis and alerting, and health-check endpoints.
• Protected production data with layered authentication and authorization, input validation, API security controls, and masking that keeps PII and tokens out of plain-text logs.
• Validated API capacity with JMeter, exposing bottlenecks, connection-pool constraints, and memory issues before they reached users.
• Built Kafka-based processing flows using deliberate partitioning and delivery-semantics strategies for dependable asynchronous workloads.
• Automated repetitive business and team workflows, reducing manual handoffs and creating more consistent, auditable execution paths.
• Developed Amazon Q AI agents with production-oriented prompt practices focused on consistent and reliable workflow outputs.

• Automate deliberately: if a repeatable process is being performed manually, I look for a reliable way to automate it.
• Design for change: clear boundaries, SOLID principles, and readable code keep enterprise systems adaptable.
• Build quality in: unit, integration, regression, and load testing are part of delivery—not post-release activities.
• Secure by default: authorization, validation, API protection, and safe logging begin at design time.
• Operate what I build: observability, health checks, failure handling, and rollback paths are core features.

📊 GitHub Overview      Statistics reflect public GitHub activity and do not represent overall professional experience.

I'm always interested in discussing enterprise Java, backend architecture, distributed systems, application modernization, and business workflow automation.

• 💼 LinkedIn
• 🧩 HackerRank
• 📧 saurabhsds13@gmail.com