Pressnave Kiruparaj

Building a banking and fintech platform supporting fiat and crypto operations with reactive, secure, independently deployable services.

• Designed WebFlux microservices with CQRS across JPA and R2DBC workloads.
• Built a financial tick ETL pipeline processing 50K+ events per second into multi-timeframe OHLC data.
• Implemented KYC, virtual IBAN, transaction processing, Kafka workflows, and secure API integrations.

Led the migration of a classified marketplace from a monolith to modular microservices while adding AI, lead-management, and observability capabilities.

• Reduced dynamic UI metadata retrieval latency by 80% using SQLite and S3.
• Improved backend performance by 40% through async validation, Redis events, and optimized SQL views.
• Delivered RAG services, remote and STDIO MCP servers, gRPC orchestration, and a Go rollout tool.

Developed secure, real-time marketplace services and production RAG pipelines with scalable Kubernetes delivery.

• Built hierarchical RBAC, token propagation, RabbitMQ lifecycle workflows, and WebSocket chat and bidding.
• Improved PostgreSQL and Redis throughput by 35%.
• Evaluated LLM grounding and hallucinations, then improved responses through retrieval and prompt optimization.

Led microservice modernization and real-time data engineering for ERP, logistics, ML features, and AI-enabled enterprise workflows.

• Built Kafka and Flink pipelines for fleet GPS, delivery events, enrichment, aggregation, and feature generation.
• Delivered stateful exactly-once stream processing and Scala/PySpark batch workloads.
• Mentored engineers and guided architecture, reviews, delivery practices, and multi-agent AI integrations.

Built event-driven services, real-time WebSocket pipelines, event sourcing, reusable interfaces, and automated delivery workflows.

• Implemented Kafka and RabbitMQ asynchronous workflows.
• Optimized SQL schemas and deployed microservices through Docker and Kubernetes.

Built a highly available WebRTC communications platform with AI-assisted intent detection and ticket routing.

• Scaled Go WebSocket signaling horizontally with RabbitMQ fanout.
• Delivered React clients, LangGraph automation, and Kubernetes HA deployments.

Engineered a real-time fleet platform ingesting high-volume GPS telemetry from embedded devices over raw TCP.

• Built Netty/NIO TCP servers and binary protocol parsers for thousands of concurrent device connections.
• Delivered geofencing, trip reconstruction, alerts, dashboards, and low-latency storage patterns.

Developed real-time IoT and computer-vision systems connecting cloud applications with edge hardware.

• Built MQTT pipelines, React dashboards, and low-latency Java clients.
• Integrated TensorFlow, OpenCV, FastAPI, and NVIDIA Jetson for real-time control and inference.