Professional Journey

Zoho Corporation

Chennai, IN | June 2023 – Present

Orchestration & Scalability: Designed a Kubernetes-based orchestration layer for AI/ML workloads—cutting deployment time by 40%, reducing manual config by 60%, and boosting productivity by 50%.
Infrastructure & Deployment: Migrated a monolithic service to microservices with Kubernetes—achieving auto-scaling, rolling updates, 3× scalability improvement, and 90% downtime reduction.
High-Performance Distributed Storage: Built a gRPC-based distributed database (with cache layers) supporting 1M+ requests/day—featuring similarity caching, TTL, pre-fetching, and write-back, reducing cache misses by 35% and improving read speeds by 70%.
Optimized Docker Workflows: Adopted multi-stage Docker builds—shrinking image size by 40%, cutting build time from 10 to 4 minutes, and improving deployment efficiency 3×.
Message Queue Orchestration: Developed a dynamic RabbitMQ consumer manager that scaled based on queue metrics—eliminated OOMs, improved throughput by 2.5×, and sustained 99.9% delivery reliability.

Chennai, IN | August 2022 – May 2023

ML Model Optimization: Led ONNX adoption for 150+ models, resolving dependency issues and reducing deployment time by 30%, improving response times by 25%.
Low-Latency Model Serving: Converted models to ONNX, cutting inference latency from 150ms to 100ms for 200K+ requests/day while maintaining 99.8% accuracy.
In-Browser ML Inference: Enabled client-side ML by loading ONNX models in browsers—reducing server load by 50% and doubling response speeds.
Efficient In-Memory Caching: Built a high-performance LRU cache that cut datastore queries by 40%, reduced latency by over 50%, and sustained a 99.9% hit rate across 2M+ lookups/day.