Designed and enhanced a distributed traffic probing system (Byomkesh) to monitor inter-DC latency, improving visibility into TP gateway performance across GCP/FCP environments and reducing debugging time for latency issues

Implemented dynamic routing and load balancing logic for TP gateways, improving scalability and optimizing latency accuracy across p95 and p99 metrics, reducing latency variance by 10-15% across network paths

Developed XDP/eBPF-based high-performance packet processing solutions in GCP for IP decapsulation, overcoming cloud networking limitations, optimizing packet flow while handling high-throughput traffic (10k+ packets/sec)

Built and enhanced Kubernetes-based TP controller to handle pod lifecycle events, ensuring high availability and maintaining accurate service whitelisting, resolving inconsistencies caused by API-server sync issues in distributed environments

Automated infrastructure provisioning and deployment using Ansible, reducing manual deployment effort by more than 30% and improving release turnaround time

Enhanced system reliability and observability by integrating monitoring and alerting using Prometheus and Grafana, proactively detecting failures in production systems, reducing incident response time

Managed production systems at scale during high-traffic events like Big Billion Days, ensuring system stability, fault tolerance, and horizontal scalability under peak load