Backend engineer for Samsung TV Plus Streaming Service.

• Lead engineer for development of fitness application which included backend services, mobile, tv, and web applications.• Managed a team of more than 10 remote engineers in US, Israel, and South America.• Provisioned and configured infrastructure (Google Kubernetes Engine, Google Cloud DNS, Google Cloud Storage) in Google Cloud Platform.• Wrote Kubernetes manifests using Kustomize to configure and deploy microservice stack to multiple environments.• Built devops systems and processes to automate code builds and deployment of builds to multiple Kubernetes clusters. Integrated CircleCI and Cloud Build build status with Slack to provide feedback to engineers upon committing code.• Integrated GraphQL stack with Apollo Studio to monitor, analyze, and optimize web service performance.• Developed NodeJS backend services and Google Cloud Functions to support uploading and transcoding of video content. Video transcoding was performed by building an integration to a third-party service.• Built initial creator platform POC to allow fitness trainers to upload, annotate, and publish fitness videos.

• Architected, developed, and deployed GraphQL-based microservices on AWS EKS.• Designed AWS DynamoDB "schema" providing performant and scalable queries.• Evaluated and recommended technologies for use in projects' microservice stack.• Developed Kustomize structure/resources and build scripts to support re-use of Kubernetes resource definitions in different environments (local Minikube development, AWS EKS staging/production)• Developed GraphQL middleware to provide IAM policy-enforced row-level access to DynamoDB.

Architected, developed, deployed, and managed systems comprising the Workpop platform.• Architected and implemented scalable backend for real-time messaging service built on MongoDB, Elasticsearch, Redis, GraphQL, and node.js.• Introduced Google Pub/Sub into Workpop's architecture to decouple services and provide reliable inter-process message delivery.• Developed infrastructure to enable efficient deployment of microservices to both AWS Elastic Container Service (ECS) and Google Cloud (via Kubernetes).• Developed libraries to instrument response times and invocation counts of GraphQL microservices to an InfluxDB instance.• Refactored MongoDB schemas to increase query/write performance.• Improved Workpop Job Search relevancy and performance by converting previously implemented MongoDB search to use ElasticSearch. Tuned ElasticSearch queries including support for fuzzy text search and term weighting.• Developed libraries to enable the syncing of data between MongoDB and ElasticSearch.• Integrated partner job sites including Indeed and Jobs2Careers into job search results.• Designed and developed backend for Workpop's SKU and credits system.• Developed backend for mobile (iOS/Android) in-app purchases with purchase validation through integration with Apple and Android subscription/validation APIs.

Architect for EV Cloud platform. The EV Cloud platform is a REST-based, multi-tenant platform allowing organizations to manage their electrical vehicle charging infrastructure. The EV Cloud platform is offered as Software-as-a-Service (SaaS) and is consumed by various clients including iOS and Android native apps, web-based apps, and third-party integrations.• Architected the EV Cloud platform. Constraints include providing a highly available, scalable JSON-based REST service, with pluggable integration with multiple vendors of charge stations.• Designed MongoDB NoSQL database schema to provide scalable database performance based on access patterns.• Led development of EV Cloud platform including specifying and implementing suitable technology to satisfy design requirements.• Constructed testing and deployment infrastructure which includes: source control using GitHub, automated testing using Jenkins, deployment to Amazon EC2 using Jenkins and AWS OpsWorks.• Developed the EV Connect iOS mobile application. This application has been released on the App Store.• Developed EV Cloud administrative web portal, allowing administrators to manage their EV charging infrastructure including viewing status, reporting on usage, and managing access.• Worked with GE engineering team to integrate the EV Cloud platform with their EV charge station hardware and APIs.• Built and led EV Connect development and QA team.• Implemented plugin architecture to provide extensibility, stability, and ease of feature development to the EV Cloud platform. This plugin architecture also now serves as one of the ways to package and monetize sets of features.• Developed documentation framework for JAX-RS providing generation of a web-based REST client for API endpoints.• Implemented Splunk logging infrastructure to enable the monitoring of API usage.• Developed Open Charge Point Protocol (OCPP) 1.5 Central System.

Engineer on LinkedIn Recruiter product which provides recruiters and other hiring professionals the ability to find and hire the most qualified candidates for their needs.• Engineer on profile visibility project to differentiate feature sets over different subscription levels.• Engineer of Inside Opinion feature allowing Recruiter users to get references for potential candidates.

• Lead developer of new (built from ground up) RESTful metadata service used to provide fast, scalable storage and retrieval of media-based metadata. The metadata service's goal was to provide below 2-ms average lookup time of image metadata for the millions of images uploaded and accessed on eBay each day. MongoDB and the RESTful web service were deployed over multiple datacenters. • Implemented JSON-based RESTful service using Sun Jersey's JAX-RS implementation.• Worked with eBay architects and 10gen engineers to design, implement, and deploy MongoDB schema capable of providing quick and scalable storage and retrieval of metadata.• Created and executed distributed Jmeter tests to analyze the load and performance characteristics of MongoDB and metadata service.• Evangelist for use of open-source technologies within the team.• Mentored junior engineers.

• Java developer on MediaPlatform's Primetime product, a platform allowing enterprises to upload, manage, search, and stream video content via a personalized portal. • Created RESTful JSON API for Primetime product using Spring.• Designed and developed video transcoding framework capable of encoding videos to the Adobe HTTP Dynamic Streaming protocol, as well as Apple's HTTP Live Streaming format.• Instituted Nexus and Hudson CI into software development workflow to provide feedback on build readiness as well as to provide deployment of new builds to staging and production environments.• Integrated with Splunk to build a system to track per-customer video streaming usage. Reports generated by this system could then be used to determine the viewership of webcasts as well as used for billing.• Converted Apache Ant-based build system to Maven.

• Developed server-side components for Norton Online Family, a web-based application allowing parents to monitor and restrict their child's online activities. • Developer on Symantec Security Information Manager, a network appliance used to correlate events originating from an enterprise's many security devices.• Developed in-house SKU management application using GWT (Google Web Toolkit).

• Developed Distributed Media Processor for MediaPlatform. The Distributed Media Processor is a service-based system transforming multimedia assets into formats appropriate for webcasting.• Developed real-time viewer statistics collection and reporting engine incorporating a SQL Server/Oracle database backend, JSP, and Java servlets. This system provides immediate feedback to our clients, allowing them to determine the effectiveness of their broadcast.• Developed C++-based application to encode streaming video using Windows Media SDK and Real Producer SDK.• Designed and developed distributed web services queuing and workflow architecture to asynchronously process various multi-step tasks. One specific application is the ability to encode videos into multiple formats and bandwidths, and uploading the encoded videos to appropriate streaming servers.

• Experimented with various image and video wavelet-based compression algorithms.