San Francisco Bay Area
I’m a Software Engineer at Amazon focused on building scalable distributed systems, AI infrastructure, and high-performance computing platforms. Previously at Juniper Networks and Qualcomm, I worked on networking infrastructure, routing systems, telemetry platforms, and parallel computing solutions for large-scale environments. My core interests include distributed systems, Agentic AI, AI infrastructure, parallel computing, networking systems, and scalable backend platforms. I enjoy solving complex infrastructure problems and building reliable systems that operate efficiently at scale.
• Architected a high-scale, event-driven benefits processing system using AWS SNS/SQS and Python, handling millions of events per day with sub-second eligibility decisions; implemented a ledger-based consistency model, driving $8M+ in annual cost savings and improving processing latency by 40%+. • Built extensible rule-evaluation pipelines and prototyped intelligent agents to orchestrate API-driven workflows, enabling automated ingestion and decisioning; reduced manual operations by 70%+ and improved end-to-end processing turnaround by 3x. • Engineered a distributed file ingestion system with chunked uploads, parallel validation through rule engines, and real-time state tracking; scaled to TB-scale daily ingestion and thousands of concurrent uploads, improving throughput by 5x while ensuring reliability and observability.
◦ Built an event-driven telemetry pipeline to collect routing metrics and store them in a time-series database, enabling anomaly detection algorithms to proactively identify and auto-remediate network issues while improving reliability and reducing manual intervention. ◦ Contributed to master-slave configuration management for a session-aware data plane, supporting traffic flow orchestration and high availability within SD-WAN architecture. ◦ Designed a local router override utility that allowed routers to operate independently from remote servers, ensuring uninterrupted connectivity for locally connected devices during server-side failures or outages.
Developed a console to onboard digital content subscriptions.
◦ Developed low-level user-mode components for automotive platforms (Android/QNX), enabling reliable execution of GPU workloads and managing resource interaction between applications and hardware. ◦ Built profiling and instrumentation tools to capture CPU/GPU utilization, memory bandwidth, and frame timing metrics; identified pipeline bottlenecks and implemented cache-aware memory optimizations, improving throughput by 20% and frame latency by 15%. ◦ Engineered debugging and validation frameworks for Snapdragon-based platforms, enabling reliable system initialization and improving observability of hardware–software interactions through API interception and diagnostics tooling.
Worked as a Developer in Media Domain for Infotainment systems