Aug 2020 – Mar 2022 · 1 yr 7 mos

• Project:
• Add SASL/PLAIN support to Self-Managed Kafka Event Source:
• Designed and implemented the workflow to add SASL/PLAIN Authentication support for Kafka Clusters.
• Developed Cloudformation-init scripts to deploy EC2-based SASL/PLAIN with TLS Encryption KafkaClusters worldwide for integration and end-to-end testing.
• Monitored SASL/PLAIN workflow and test infrastructure by setting up monitors, alarms and performance metrics using Cloudwatch.
• Handled operations by helping customers to onboard their Kafka clusters with AWS Lambda.
• Tools: Java,Cloudformation, Shell-Scripting, AWS EC2, Kafka, Cloudwatch, DynamoDB, S3, Route53
• Add SASL/GSSAPI (Kerberos) support to Self-Managed Kafka Event Source:
• Implemented a running prototype to set up Kerberos Authentication for Kafka Event Source.
• Designed detailed APIs and workfl ows for incorporating Kerberos Authentication with Lambda.
• Tools: Java, Kafka, Kerberos, Cloudformation, AWS EC2, Cloudwatch, S3, Shell Scripting, Route53
• Load Testing for Amazon MQ and Kafka Event sources:
• Implemented a TPS load-generator for stress testing the polling system to identify bugs and collect performance metrics for monitoring the feature launch.
• Tools: Java, Rate-Limiter, Cloudwatch, Amazon MQ

Sep 2019 – Dec 2019 · 3 mos · Santa Cruz, California

• Computer Systems and Assembly Language

Jun 2019 – Sep 2019 · 3 mos · Ann Arbor, Michigan

• Project:
• Performance Analysis of Microservice-based Application Using Queuing Theory:
• Applied Layered Queuing Network (LQN) simulation model to the microservice-based platform to collect system’s performance metrics (throughput, utilization, and response time). The obtained performance model helped in identifying architectural bottlenecks and system’s response latency improved by alleviating these bottlenecks.

Sep 2018 – Jun 2020 · 1 yr 9 mos · Santa Cruz, California

• Project:
• Cooperative-LSM: Designed and implemented C-LSM trees by deconstructing monolithic LSM trees into three components - leader, compactor, and backup. Leader serves to write traffic, compactors handle compaction operation, and backup provides read availability. C-LSM maintains a consistent read/write performance for applications with dynamic workloads where read/write access patterns are unpredictable, and no clear partitioning is feasible.

Sep 2018 – Dec 2018 · 3 mos · Santa Cruz, California

• Introduction to Operating Systems

Jun 2014 – Jun 2018 · 4 yrs · Gurgaon, India · On-site

• Projects:
• Platform Independent Cache Services:
• Designed, developed, tested (Visual VM and JMX) and deployed platform-independent cache services for efficient processing and fast retrieval of data from the underlying database in a real-time client-server environment which helped in a smooth transition to AWS.
• Database Migrated from Cassandra/MySQL to Aerospike:
• Developed batch & real-time scrubbers (full and incremental) to modify data in the NoSQL database (Aerospike) by modeling complex data structures that improved websites response time by 20%.
• Database Failed Events Reprocessing Framework:
• Designed and implemented Failed Events framework from scratch by pushing real-time database failed events to HBase via Apache Kafka for further reprocessing of events using Apache Storm topologies that reduced stale data issues on the website by 30%.