Working on resilient scale-out grid platform for data synchronization across any topology or database platform on any cloud

Worked on developing several backend services using Java 8, Play Framework, Akka

IoT Team:

Brief description: Working on highly scalable secure cloud component of IoT infrastructure, used for IoT devices representation and relations storage, capable of secure interoperation with other services by means of different protocols (MQTT, HTTPS, CoAP)

• Engineered the design of a scalable system taking into account all of the major requirements

• Built a scalable RESTful application used as device representation storage mechanism using Spring Boot as the back-bone, Spring Data for ORM, Neo4j and Couchbase as a polyglot database solution (to be able to represent devices according to specific industry standards)

• Designed and developed a JWT based authentication service application that is used by a variety of other services as well as users for authentication and authorization, using Auth0 as underlying JWT provider and user password/ username storage

• Developed several microservices using Spring Boot: bridge microservice between MQTT and Kafka, api gateway for telemetry data consumption

• Performed tuneup of applications servlet containers (Jetty 9): customized requests queue size, number of threads etc., safe guarded calls to external services using Hystrix circuit breaker

• Dockerized all developed applications

• Architected the applications infrastructure based on SOA principles and deployed the one in AWS cloud. Used AWS ECS (Elastic Container Service) and other AWS services: ECR, ELB, Route53, EC2, S3, ApiGateway to implement a scalable architecture without downtime during deployments by utilizing blue-green deployment pattern

• Codified the entire infrastructure using Cloudformation to be able to create new environments

• Setup deployment procedures and release flow (Git Flow) using CircleCi as a CI tool

• Wrote multiple utilities scripts using bash, Python

• Performed load testing of the applications, while monitoring the JVM's using jvisualvm and making adjustments if needed

• Wrote documentation in Apiary for developed RESTful applications

WPM (WebPerformanceManagement) Load Testing application Team:

Brief description: Worked on distributed, highly scalable (up to several thousand servers) multi-cloud (AWS, GCE, Rackspace) based system used for generating traffic to customers' back-end systems/ websites

• Responsible for new features development and sustaining of the WPM Load Testing cloud based application, managing/ improving the infrastructure with a potential capacity of several thousand virtual machines in 3 different clouds (AWS, Rackspace, Google Compute Engine) that are launched and configured on demand for customers load tests

• Integrated Google Compute Engine cloud in Load Testing Application through reuse/ extension of existing code

• Implemented Hierarchical Accounts architecture in Load Testing, Scripting, Alerting applications under WPM

• Performed various Linux OS/ JVM level tune up’s, migration, upgrades of Load Testing cloud servers to extend and improve the productivity of the ones

• Worked on rewriting (using Python), consolidating and integrating deploy scripts with CI

• Assisted Professional Services/ customer support teams in various ways by looking into customer issues, quickly delivering solutions to resolve customer issues in production

• Improved Rackspace/ Google Cloud agents launching mechanism by utilizing concurrency concepts

• Set up logs reporting to Loggly from cloud based virtual machines to early identify potential issues in case virtual machines failed to launch normally as well as for general usage

• Worked on fixing various bugs/ issues in WPM Load Testing application

• Improved the functionality of some internal components by integrating Redis in-memory database into Load Testing application

• Participated in code reviews and application architecture design meetings

• Performed code refactoring/ improvement through design patterns implementation, code reuse