Data & ML Engineer, null, 2021 - present
Gainy is a personalized asset management app for retail investors.
As a part of the engineering team at Gainy, I was responsible for many data-driven aspects of the product:
- Personalised match score based on users' interest and investment profile
- Recommendations of stocks and stock collections
- Smart in-app search
- Automatic improvement of stock cards (e.g. automatic refinement of industries based on stock descriptions)
Technically speaking, I prototyped and built:
- Data processing infrastructure and pipelines (AWS, Terraform, Airflow, DBT)
- Integration of multiple sources of stock market data
- Production-level machine learning models for NLP and recommendation system (Python, Tensorflow, Scikit-learn, NLTK, MLflow)
- In-app search (based on Algolia)
Solution Architect, EPAM Systems, 2018 - 2021
- Designed and developed modern and traditional Data & Analytics platforms, following Data mesh, Data fabric, Data lake and Data
warehouse concepts and patterns; experienced with Lambda and Kappa architectures.
- Was responsible for technical aspects of multiple projects, including customer data platform and customer analytics (ML-based
personalisation and customer scoring) in Retail; data warehouse cloud migration in Automotive; core platform capabilities (data
integration & orchestration, streaming data, data observability) and data products in Retail and Finance.
- Built solutions on Amazon Web Services (AWS) and Microsoft Azure with a mix of platform-as-a-service (S3, EMR, HDInsights, Cosmos
DB, Snowflake, Databricks) and open-source technologies (Elasticsearch, Cassandra, Kafka, Airflow, Hive, Presto).
- Worked on various sized projects, starting from an MVP with 3 - 4 engineers up to leading a team of architects on a large cloud
migration with 100+ people totally; experienced with Agile methodologies.
- Architected and prototyped frameworks for data processing, data observability and machine learning operationalisation in Scala and
Python; approached data management and data governance automation through comprehensive metadata management.
Lead Data Engineer, EPAM System, 2017 - 2018
- Participated in multiple big data projects, e.g. built core data platform capabilities (data processing and orchestration) and various
data products in Retail; implemented a highly-performant recommendation engine in Telecom.
- Implemented highly-loaded web services (Java, Spring Framework, big data processing pipelines (Apache Spark, Scala and Python)
and databases (Cassandra, Apache Ignite) with strong focus on engineering excellence.
- Worked closely with Data science team on optimisation and productisation of machine learning algorithms using Apache Spark MLib.
- Had experience in team leadership, including hiring and project interviews, mentoring, performance reviews; evolving teams from
"forming" to "performing" phase.
Lead Software Engineer, Reltio, 2015 - 2017
- Designed, prototyped and developed an analytical platform within Reltio Cloud, designed to seamlessly integrate data to quickly
enable predictive modeling and reporting (with Databricks).
- As the technical lead of the analytical platform, was responsible for key technical decisions, including event-based data
synchronisation (with Amazon Kinesis), data storage optimisation (on Amazon S3), data pipelines (Apache Spark), etc.
- Took responsibility for overall team management, including hiring interviews, mentoring and performance reviews; as well as
communication with the product owner and end users.
- Re-engineered the existing data matching algorithms by using a distributed computation engine (Apache Spark) and, therefore,
increased the performance by 10x and enabled fuzzy matching on big data scale (billions of records, tens of terabytes).
Senior Software Engineer, Reltio, 2013 - 2015
- Worked as part of the team responsible for Reltio Cloud platform, which combines modern master data management with data-driven
- Successfully implemented new features in Java and Spring-based web services, including attribute-based data access, data
import/export, matching functionality, etc.
- Fully owned the part of the part of the platform related to record matching (i.e. fuzzy data deduplication), including design and
implementation of new functionality, database performance optimisation (Cassandra), etc.
- Closely worked with the data science team on machine learning algorithms in matching, including optimisation and productisation of
the proposed approaches.
Software Engineer, Axiom SL, 2012 - 2013
- Was the part of a backend team, which worked on a regulatory tool to deliver risk aggregation, workflow automation, validation and
audit functionality, etc.
- Implemented multiple new features, as well as internal optimisations of client-server communication.
Junior Software Engineer, Yandex, 2011 - 2012
- Participated in an internal project aimed to improve the quality of data about Russian organisations (in Yandex.Map and other
services) by applying machine learning techniques to web data extraction.
- Worked closely with data scientists to propose and implement a novel approach to web information extraction based on visual
representation of web pages.