Maria Castro

Architected smart data lake using Apache Iceberg and DuckDB, reducing raw data storage from 70TB to optimized Parquet format on S3. Implemented cost optimization strategy and maintained ML-based property matching models across 55M+ listings.

Designed and implemented ML-based entity matching pipelines for property deduplication across OTA platforms. Built data integration pipelines consolidating 2.2T+ calendar records and real-time market intelligence infrastructure.

Built enterprise ETL solutions using Apache Airflow integrating diverse data sources including insurance provider files, Salesforce, NetSuite, and call center systems. Designed schema architecture for insurance business logic and analytics infrastructure for agent performance.

Designed ETL processes using Azure Data Factory for multi-source data integration and built pipelines to ingest data into Azure Data Lake Storage. Loaded transformed data into Synapse and optimized SQL for performance.

Delivered end-to-end data solutions across diverse client projects including ETL pipelines, data architectures, and BI dashboards. Built Azure Data Factory pipelines and configured Key Vault-backed linked services for secure credential management.

Developed complex data analysis reports and dashboards for financial and operational metrics using Lumina, SAP BusinessObjects, and Crystal Reports. Supported SAP HCM migration and system integration, and received an award for developing an interface process between accounting and HCM systems.

Master's Degree in Information Engineering from Universidad de Los Andes, Bogotá, Colombia, from 2017 to 2020.

Bachelor's Degree in Systems Engineering from Universidad Distrital Francisco José de Caldas, Bogotá, Colombia, from 2002 to 2010.