Aug 2023 – Jul 2024 · 11 mos · Gurugram, Haryana, India · On-site

• 1. Designed and directed automated EKS and add-on upgrades using ClusterAPI and ArgoCD, cutting upgrade times from months to days. Contributed to the CNCF ClusterAPI project, enhancing its functionality and reliability and to support Gemini's usecases.
• 2. Structured platform documentation in Confluence, guiding effective Jira usage.
• 3. Resolved company-wide autoscaling issues with gRPC services by addressing persistent connection challenges, optimizing load balancing.
• 4. Designed and directed unified PKI and secrets store, streamlining company methods.
• 5. Contributed to the data center to AWS migration effort.
• 6. Calculated the exact Multicast UDP latency numbers in AWS over the transit gateway, earning validation and appreciation from key stakeholders after initial skepticism. Proposed an innovative eBPF-based multicast with unicast approach, reducing hop latency from 150 ms to 10ms in AWS, achieving latency comparable to that in a data center.
• 7. Spearheaded the design and implementation of a network devices syslog logging setup in AWS, guiding a team to simplify the process and reduce the number of hops, resulting in significant cost savings.
• 8. Educated cross-functional teams on using Envoy for traffic management and planned the integration of multiple tools into a single tool, Istio.

Jun 2022 – May 2023 · 11 mos · Gurugram, Haryana, India

• 1. Led service mesh infrastructure design using Envoy and Kuma, facilitating service communication, load testing, training, and canary deployments.
• 2. Achieved a 10% reduction in AWS costs by optimizing the Golang garbage collector.
• 3. Designed and led the development of in-house high-availability monitoring systems using Prometheus and Grafana, resulting in improved incident response times and increased system availability.
• 4. Reduced 60% Zomato's observability infrastructure metrics cost by migrating from paid option to an in-house high-availability setup.

Oct 2020 – Jun 2022 · 1 yr 8 mos · Gurugram, Haryana, India

• Part of Site Reliability team

Oct 2019 – Oct 2020 · 1 yr · Gurugram, Haryana, India

• Part of Site Reliability team

Apr 2019 – Sep 2019 · 5 mos · Gurgaon, Haryana, India

• 1. Created a Proof of Concept for a Server-Subscription based model to automate the QA pipeline, addressing the challenge of hardware-specific NIC dependencies without relying on Kubernetes and its added operators.

Jul 2016 – Apr 2019 · 2 yrs 9 mos

• 1. Acted as a key interface for Simulated test infrastructure, developing a C tool using Linux cgroups for resource tracing and sandbox testing. The tool resulted in a 50X increase in server productivity, addressed the long-running OOM problem, and introduced resource capping to control completion time inconsistency and unknown timeouts in tests.
• 2. Successfully executed a Proof of Concept to address the long-running GCC virtual memory exhaustion issue in 32-bit Arista operating systems, with the assistance of one overseas intern.
• ● Implemented Kubernetes for container orchestration (automated deployment, scaling, operations), developed Ansible playbooks for cluster state preservation and configuration, and migrated internal services for improved CI/CD processes.
• ● Developed a customizable golang health check tool for internal services, providing combined status of multiple endpoints (HTTP/TCP) with graphical representation on a newly created status page.

Jan 2016 – Jun 2016 · 5 mos

• Worked on Dockerizing the Services at Arista.

Nov 2012 – Jun 2015 · 2 yrs 7 mos · Allahabad Area, India

• Configured/Designed Servers/Services and maintained internet/intranet services at IIIT Allahabad.