Ihor (Igor) Novoseltsev

Project: Migration of ETL data workloads from AWS to GCP - Led the transition of transactional data from AWS to GCP, resulting in a 15% reduction in computational expenses and an 8x speed up in transformations. - Designed a multi-layer data management solution including a data lake, data warehouse, and monitoring components for OLTP and OLAP workloads. - Migrated OLTP workload from MongoDB to PostgreSQL and implemented a data warehouse layer using BigQuery, improving access to data for analysts and increasing visibility of data flows. - Enhanced monitoring capabilities through the introduction of data management solutions. - Decomposed a monolithic application by switching endpoints to separate microservices, leading to improved scalability, maintainability, and development velocity.

Project: Forecasting pipeline to compute future sales and volumes. - Designed and implemented a machine learning pipeline to process and analyze large volumes of data, with the goal of predicting future sales and volumes. - Used Apache Airflow to design and productionize ML pipelines for deploying models, monitoring performance, and handling errors, resulting a significant increase in prediction coverage by up to 92%. - Participated in the design and coding of a REST API to serve prediction data to the web application, and identified and resolved bottlenecks in the system, leading to a 50% reduction in API call latency. - Configured and set up a Spark cluster for distributed computation of machine learning models, implementing a production configuration that resulted in up to 3x faster computation and a more than 65% reduction in computational costs compared to the initial setup.

Project: AWS to GCP migration. - Participated in a project to migrate an enterprise-level infrastructure from AWS to GCP, including planning, design, and implementation. - Contributed to the implementation of a new infrastructure on Google Cloud Platform (GCP). - Worked on improving the CI/CD process, streamlining code deployment and testing procedures. - Collaborated in decomposing monolithic applications into microservices to support cloud-native architecture. - Provided expertise and guidance on GCP cloud services and best practices to ensure a successful migration. - Worked closely with other team members and stakeholders to ensure timely delivery and effective communication throughout the migration process.

Real-Time Dashboard: - Led the implementation of a real-time dashboard as a single-page application (SPA) using React and Socket.io, improving data refresh times by 4x on the frontend. - Designed and implemented real-time communication channels between web clients, server, and the strategy manager to enable faster decision-making. - Gathered requirements, estimated tasks, and coordinated implementation efforts to ensure timely and high-quality delivery. - Technologies used: FastAPI, Socket.io, React, Redux, and Redux Saga.

Project: Test Runner Tool. Designed and maintained services and applications to manage an internal fleet and support automated systems testing. - Developed expertise in utilizing tools such as Grafana, Prometheus, and Elasticsearch to gain deep insights into system performance and health. Implemented Grafana dashboards based on Prometheus metrics, which enable increased visibility and streamlined monitoring. This has led to an increase in the efficiency of cluster resource utilization by up to 30% - Designed and implemented a Test Runner Tool for multi-system ecosystems, using Docker containers and an autoscaling group to reduce costs by 65% and increase run speeds by 3x. - Utilized BigQuery as a data warehouse for historical data ingestion and further analysis, enabling efficient data querying and analysis on large datasets. - Led the development of a POC to migrate an AngularJS SPA to React and successfully managed the implementation and migration phase while preparing thorough documentation for future development and team onboarding. - Streamlined the development process through effective planning and communication. - Developed and implemented solutions to improve efficiency and productivity.

Project: Artist Dashboard. Developed and designed the Artist Dashboard, a rich internet application for tracking musicians' sales and metrics. - Developed Single Page Application (SPA) architecture using AngularJS - Created AngularJS components for the application - Utilized node.js for middle layer proxying, caching, and authentication

Front-End Component Library: - Designed and developed a reusable components library for web UI projects - Improved efficiency and maintainability by creating a UI Components API with a focus on inheritance and customization - Integrated the library into multiple web user interfaces - Implemented thorough testing to ensure reliability and consistency - The library was successfully reused by other teams, saving time and resources in future development

OsCommerce / TYPO3 / WordPress / BuddyPress / Yii - integration, extensions development. MySQL, Sphinx full text search engine.

OsCommerce / TYPO3 - integration, extensions development.

Design, development and support of corporate site.

Thesis - Methods and means for recognizing the changes of the object properties through images based on artificial neural networks.