- RCS Channel
- Implemented an event-driven asynchronous message delivery service for hypermedia messages at scale
- Developed two microservices, a sender service for AWS SQS message consumption and a tracker service with REST API endpoints for delivery receipts
- Utilized Docker, Kubernetes, GitLab, and Argo CD for seamless deployment and CI/CD pipeline management
- Achieved 30k/min throughput with optimized resource utilization, ensuring robust performance
- Campaign Stats Druid
- Developed an event-driven system to asynchronously process CRM data from a Kafka topic
- Persist data into Redis, Apache Druid for low latency access for CRM dashboard
- Achieved an ingestion rate of 50,000 events per minute with minimal resource usage
- Summarized Reports
- Developed a multi-threaded report generation service for on-demand and scheduled CSV reports
- Utilized Apache Druid for data retrieval with efficient and low-latency queries
- Implemented batching and streaming for better efficiency of the system
- Automated report generation in various formats as per the user specifications
- Campaign Stats Query Server
- Built REST API-based query generation service interfacing with Redis, Druid, Google BigQuery, and MongoDB
- Used mustache templating engine to generate queries based on user input with pre existing template queries
- Optimized Druid and BigQuery queries to reduce cost, minimize data volume, and ensure low-latency data