Experienced in designing large scale distributed systems.

Started as the tech lead for the Discover Serving team and worked on projects ranging from launching localized content in the MENA region, rewriting the serving endpoint for better performance & rapid experimentation, cutting down costs by nearly half and introducing a generic framework for serving recommended content based on asynchronous & non-blocking principles.

Led a team of engineers in designing and implementing the backend service to serve Friend Feed - a combination of chat & stories content to the users, during a major app redesign.

Conceived the idea for a prototype of the Friend Feed service, optimized for end-to-end speed, performance and cost. It is a complete rewrite of the entire Friend Feed service. Led the engineering effort for the prototype and worked with a team of engineers to implement and scale the backend service.

Currently working on the Ads Delivery team on designing scalable & efficient architectures for Ad & Product Serving systems.

1. Led the design & implementation of scaling the Dynamic Product Ads infrastructure. It was able to scale the products from < 10M previously to 100M & beyond. In the process, designed the platform for a Product Metadata data service which gets updates through a delta sync mechanism.

2. Designed & led the implementation of the Ads selection service which introduced using smarter indices to speed up index lookup for Ads retrieval. Reduced the latency as well as improved the throughput of the filtering stage.

3. Infra cost POC for the Ad Delivery ecosystem spanning over a 100 cost subsystems.

4. POC for the Ad Delivery team's roadmap.

Working in the Ad serving team at Twitter