Seller Data Platfrom team• Taking part in building a new department's Hadoop-based DWH with data management and version control implemented using the Apache Iceberg catalog, and data access provided through the Trino engine.• Building new ETL pipelines orchestrated in Apache Airflow using PySpark jobs for RAW/ODS/DM layers fulfillment from different sources (Clickhouse, Greenplum, Postgres, S3). • Ingested pipeline for loading and storing streaming data from Apache Kafka. Implemented service with PySpark that stores data from raw messages to ods layer table regardless of message format and data schema.

Data&BI team (as freelance employee)• Worked on implementing data mart and view layers in Snowflake for BI team. Contributed to automating analytics team manually-run python scripts by transforming their data calculation process into an automated pipeline that are orchestrated via Stonebranch and run with Azure Data Factory processes.• Maintained existing data pipelines in Informatica ICS for data transformation and load to Snowflake stage layer from Azure Data Lake.

HR-research team• Developed processing of big data from HR platform clickstream service. Reduced metrics build time by 5 times with query optimization. Performed data mart calculation speed-up orchestrated via Airflow.• As a member of R&D team, took part in multiple data-driven researches such as identification of top factors that affect employee's flow from company. Here also provided research on referral program that allowed to adjust approaches to increase internal hiring. Took part as data engineer in A/B tests - implemented ETL with PySpark from various sources like Greenplum or Postgres for creating data marts for data visualization in Superset. • Held huge amount of daily meetings with business unit to form data-requirements, elaborate details and present results.

Internal Audit Department • As a member of internal audit department team took part in developing algorithms for determining fraudulent operations and stuck bank transactions (SQL, PySpark, Hadoop).• Was responsible for maintaining audit's Greenplum cluster. Performed implementation of algorithms for identifying irrelevant and outdated data.• Conducted validation tests for identifying downgraded ML models that are to be removed from production due to it's irrelevance.