Daler Bakhriev

Experience

• 

  Technoserve

  Feb 2018 - Aug 2018

  Data scientist

  -Developed a model of a recommendation system for the Aeroflot website as part of a Big Data implementation project-Developed several models of the film recommendation system using Spark ML-Built a predictive model to predict the value of the balance of electricity flows

• 

  X5 Group

  Aug 2018 - May 2022

  -Scaled features calculation from 600 up to 17,500 brick and mortar stores using Apache Spark-Optimized Spark applications execution time from 3 days to 6 hours-Сonfigured monitoring and alerts using Zabbix, Prometheus, Grafana -Built models for calculating key indicators for optimization of the assortment matrix in more than 15,000 brick and mortar stores-Integrated my own analytical solutions into the product backend using microservice architecture-Rapidly prototyped services for demonstration

  • Big Data Engineer

    Oct 2020 - May 2022

  • Data scientist

    Aug 2018 - Sept 2020
• 

  InDrive

  May 2022 - now

  Senior Data Engineer

  -Designed and introduced streaming pipelines using technologies such as Debezium, Apache Kafka, Google Pub/Sub, Google DataFlow, and BigQuery to capture and process over 750 GiB of real-time data per day.-Developed and implemented custom data quality tools to ensure the accuracy and reliability of data processed in the core data engineering platform, resulting in approximately 20% improvement in data consistency.-Integrated logging systems from Google Dataproc Spark jobs with Apache Airflow logs in Google Cloud Composer and GKE to improve system monitoring and debugging capabilities, resulting in more efficient issue resolution.-Designed and implemented a pipeline to transform data from tables with logs of changes streamed from Debezium into actual data tables, enabling data analysts to generate reports 4 times faster.-Developed a key Python module for ETL processes into a standalone framework distributed as a Python library and facilitated its adoption within the company-Optimized and refactored the Python ETL framework, achieving a performance increase of over 40 times-Enhanced the framework to operate with significantly improved security Show less