Sep 2021 – Present · 4 yrs 6 mos · Berlin, Germany

• At DeliveryHero, I help customers discover better restaurant and grocery products by building search and recommendation systems.
• Worked on the Favourites API, a multi-region system that handles requests for customer favourite restaurants at a rate of 90K RPM.
• Designed and led the work of syncing favourites data between clusters and into BigQuery, for example allowing data from Europe to be available in Asia for failover, and in BigQuery for analysis.
• This was done using Springboot, Google Pub/Sub and PostgreSQL.
• Worked on the Swimlanes API, a multi-region system that gets all restaurants available in the user's location and groups them in category lanes at a rate of 200K RPM.
• Integrated Recommendation Systems with Swimlanes API to improve conversion rate.
• Worked on infrastructure setup using Google Cloud and Terraform.
• Optimised cloud costs by 30% through efficient resource optimisation.
• Created dashboards in DataDog to display Latency and Request information for our system.
• Improved customer conversion rate (CVR) by 20%.