Alex Ilyin Email & Phone Number

Denver, Co, Us

● Designed ELT pipelines with 5 stages of data lineage and modeling that are orchestrated by Prefect. Source data is extracted intoAWS S3, after which it is cataloged in AWS Glue and prepared for loading into Redshift. Data is modeled in dbt and exposed toLooker, where it is used to empower company-wide data literacy and analytics. Data pipelines are subject to strict review process tomake sure data adheres to standards of healthcare data privacy (HIPAA and PHI restrictions).● Revamped the Data Engineering team’s CICD process using custom Github actions. Proposed changes to production code aretested in AWS ECR environments to ensure new changes pass necessary validation before going live.● Spearheaded project to design Salesforce data pipelines that enabled cross functional teams to access Salesforce data, as well asreverse ETL processes that enriched Salesforce with data from various external sources and helped executives with planningsurrounding new customers/healthcare provider partners, prospective market expansion, and company budgets.● Launched new data observability tool called Elementary to monitor data model run times and test results. The tool allowedengineers to identify which data models caused bottlenecks and to debug data validity test failures.● Partnered with Data Scientists and ML Engineers to define feature sets to be used for ML models and experimentation. Data wastransformed before being ingested to AWS using the FeatureStore SDK. Models were trained on the datasets using Sagemaker.

San Francisco, California, Us

● Built customer churn prediction model using Python packages Scipy and Sklearn. Used SQL to create feature sets and Pythonpackages Pandas, Matplotlib, and Seaborn to perform exploratory analysis along with feature engineering and selection. Oversawmodel deployment and implementation for churn prevention strategies led by Customer Success team.● Designed and implemented data models using dbt for analytics related to new feature releases, financial, and Customer Successdata. Built high-visibility Looker dashboards using newly implemented data models.● Used Python libraries such as Requests and Pandas create new data pipelines. Data was pulled from APIs and transformed forvisualization and analytics. The pipelines were orchestrated using Airflow.● Led effort to compile financial data required by investors during Envoy’s $111m Series C fundraising round. The data compiledwas praised by investors for thoroughness and accuracy.

United States, Us

● Designed, developed, and deployed automated data pipelines for clients ranging from high growth startups to established firms aspart of efforts to assist clients with creating efficient accounting practices and IPO readiness. Data pipelines integrated client’sCRM and ERP platforms such as NetSuite and Salesforce with transactional data from payment service providers.● Used Python libraries such as Requests and Pandas to connect to API endpoints, extract data, and to apply data transformations.● Cron task scheduling was used to kick off jobs and create data files, the files are picked up by the iPaaS API integration toolWorkato, which creates records in the client’s internal systems. AWS services such as S3, Athena, and Redshift were used tomaintain client data stores. New data pipelines decreased compilation times of legacy integrations by up to 30%.

Master'S Of Science, Analytics

Bachelor'S Of Science, Technology And Information Management