Data Engineer & Scientist

Odysseus Inc, a subsidiary of EPAM Systems, provides data engineering and analytics solutions for the healthcare and pharmaceutical industries, focusing on OMOP Common Data Model (CDM) integrations.

• Led the development and optimization of ETL pipelines that transformed healthcare data into the OMOP Common Data Model (CDM) for multiple global clients.
• Designed and implemented data architecture solutions to enhance scalability and maintainability across distributed environments.
• Collaborated with analysts and data scientists to ensure data quality, validation, and standardization for clinical research use cases.
• Improved existing Python ETL orchestration libraries, contributing to performance and reliability enhancements.
• Worked with Databricks, Snowflake, Spark, Hadoop, AWS, and Azure Data Factory to build automated data ingestion pipelines.
• Supported deployment processes, version control, and CI/CD workflows using Bitbucket and containerized environments.

Femtec Health is a US-based digital healthcare platform specializing in personalized women’s health solutions using AI, predictive analytics, and clinical data modeling.

• Established and led Femtec’s data science team of four, defining technical direction and mentoring data scientists.
• Engineered large-scale ETL processes to integrate and harmonize healthcare claims data from multiple US sources into the OMOP CDM.
• Designed and implemented predictive models for early detection of gynecological conditions using Spark, Redshift, and AWS EMR.
• Applied OHDSI frameworks for model training and external validation across datasets with diverse medical coding standards.
• Developed machine learning algorithms (Random Forest, Lasso, Ridge, Gradient Boosting, Naïve Bayes, kNN) and unsupervised models for clustering microbiome data.
• Built interactive dashboards and visualizations using Seaborn, Plotly, and Cufflinks for executive reporting.

Vivante Health is a digital health startup specializing in the management of chronic gastrointestinal diseases through the use of AI and predictive analytics.

• Developed predictive and classification models for gastrointestinal conditions using claims and wearable data.
• Built and optimized data pipelines in GCP and BigQuery, ensuring secure and efficient ingestion of medical records.
• Collaborated on feature engineering, model validation, and deployed Google Cloud AI- models into production systems.
• Contributed to the creation of automated clinical data pipelines using Python, SQL, and Apache Spark.

The Rockefeller University, located in New York, is a premier biomedical research institution dedicated to advancing neuroscience, biology, and physics.

• Conducted PhD research in Mathematical Neuroscience, modeling neural network dynamics using nonlinear systems and stochastic processes.
• Proposed a mathematical framework for adaptive, self-reconfigurable neural dynamics, which was published in Journal of Statistical Physics (2017).
• Developed neural network models that simulated input-dependent computations in the visual cortex and nonlinear wave propagation in the cochlea.
• Analyzed large-scale ECoG and multi-electrode array recordings, applying advanced statistical and computational modeling techniques.
• Authored multiple publications in peer-reviewed journals and contributed to theoretical neuroscience research communities.