Jul 2024 – Nov 2024 · 4 mos · United States · Remote

Sep 2022 – Jul 2024 · 1 yr 10 mos · United States · Remote

• 1. Working on low-latency and high throughput ad-tech systems which can handle 500k req/sec during peak load.
• 2. Designed and owned a feature to split ad requests between multiple SSPs based on configuration using Aerospike as a cache.
• 3. Split feature increased the operational efficiency of the system by 50 percent and increased revenue by 7 percent.
• 4. Working on optimizing and benchmarking ad-decisioning systems to support 500k requests per second during peak load.
• 5. Worked on various microservices to enhance user and content information in the ad requests.
• 6. Implemented frontend and backend for admin UI to configure rules to split traffic based on requests to change system behavior in real-time.
• 7. Designed a relational database to store various configurations for split traffic.
• 8. Designed and implemented a data pipeline for measurement of various metrics like KPI’s, impressions, fill rates, and revenue based on pixels which can handle data in ranges of 1M records/sec.
• 9. Enabled tracing in various applications using DataDog to trace live requests and alerts based on various rules

Jul 2019 – Sep 2022 · 3 yrs 2 mos · Schaumburg, Illinois, United States · Hybrid

• 1. Designed and developed ad server to increase fill rates and revenues for banner and video ads on LG
• 2. Designed and developed a search api on ott metadata catalog using elastic search, which can give the most rated, viewed and searched movies in the last x months.
• 3. Designed and implemented part of Server Side Ad Insertion system which inserts ads in real time live streams with low response time and high efficiency.
• 4. Designed and owned content management databases and relevant API’s for Server Side Ad Insertion system to store metadata required by the system like feed and content details.
• 5. Have worked on and developed end-to-end knowledge of various components in Ad Stack like SSP, DSP, Ad-Server.
• 6. Developed and owned an efficient ETL pipeline using Spark streaming to Elasticsearch and HDFS from Kafka near real time which scales up to 500k records per minute.
• 7. Designed and developed various backend API’s on Elastic data to calculate various metrics like impressions, uniques and campaign performance., etc.
• 8. Worked on designing and implementing various tools to format and process big data on a scale of terabytes using Spark.
• 9. Designed and implemented pixel processing ETL pipeline from Nginx to HDFS to Elastic search.
• 10. Worked on creating Device Graph which maps IP’s and devices for attribution and targeting purposes with over 100 Million entries.

May 2018 – Jul 2018 · 2 mos · Noida Area, India

• 1. Developed augmented reality based application for authentication using ARCore and Sceneform by generating challenges based on user data and surroundings.