Contributions:

● heavily involved in the re-architecture of the invoice compute engine which enabled adding most of the current features used for invoice generation

● built aggregate consistency, event matching, and aggregate computation on the aggregation pipeline

● integrated Druid into our aggregate compute pipeline and enabled real-time data aggregation

Tech stack: Java, Typescript, AWS, Terraform, Druid, Kafka, Kubernetes

● Designed and developed microservices within the BGP data pipeline which generates 30k+ data points per second

● Created an automatic release process with gradual deployment and canary support for the core application

● Reverse-engineered the Skype for Business client a feature in the platform to monitor edge servers in real-time

● Maintained and improved data pipelines - ETL

Tech-stack: C++17, Java, Python, Kafka, Elasticsearch, MySQL, Kubernetes, Docker, Jenkins, Airflow, Celery

Responsibilities:

● Designing and implementing applications for extraction, analysis, validation and

processing of financial trade information as part of a market-leading solution

● Adding support for new features in the SWIFT inter-banking communication protocol to

our application

● Implementing support for new regulatory standards in collaboration with worldwide

trade repository firms

Achievement highlights:

I improved the productivity of over 200 engineers by proposing and deploying an OpenGrok repository which greatly reduced the symbol search time in our very large code base.

I optimized the company’s development process by adding automated smoke tests to the source repository. This greatly reduced the number of broken builds by notifying developers of problematic commits and allowing quick fixes before the bug could stall the development process.

While at Luxoft I worked on:

● Firmware development for Avaya level 3 network switches

● Testing and maintenance of network protocol implementation for switches

● Feature development for various network protocols - SNMP, RMON

All the development was done in C/C++ on a Linux environment.