Business Intelligence Engineer

Develop scalable data models and ETL pipelines, enabling transparent and automated reporting of core metrics for a $550M book of business.
• Collaborate with Finance and Customer Experience leadership to automate all reporting of all account management metrics in Snowflake and Tableau, reducing manual reporting time by 100 hours monthly.
• Partner with Salesforce Architects to automate ETL pipeline for analyzing and identifying flawed closed deals, resulting in a 17% reduction in Revenue Operations support tickets.
• Create Python-based analytical model to extract, score, and load entities from the PitchBook platform to Salesforce, resulting in a 7% increase in demo completion rates.
• Construct Salesforce ELT pipelines using Python, Airflow, and DBT, increasing data refresh frequency by 300%.
• Implemented slow-changing dimensional modeling of key business metrics to enhance data governance and support comparative analytics solutions.
• Collaborate with software engineers to improve newsletter data accuracy by developing a Python data transformation pipeline to analyze subscriber usage and eliminate suspicious user behavior from reporting.

Developed and optimized ETL pipelines and quality assurance functions to enhance forecasting and dashboard accuracy.
• Automated 14 ETL pipelines in Airflow to transform, test, and blend diverse data sources, leading to a 40% enhancement in the accuracy of the global vaccine supply model.
• Collaborated with data scientists to identify modeling bottlenecks and implemented custom data warehouse tasks, resulting in a 6-hour monthly reduction in manual data entry time.
• Utilized Python to blend data from public and confidential sources, creating a Monte-Carlo simulation for precise estimation and forecasting of vaccine demand in private markets.

• Developed Python scripts to identify discrepancies across data sources, increasing supply forecast accuracy by 15%.
• Engineered backend database design of the UNICEF Vaccine Market Dashboard, resulting in improved data accuracy and a reduction in downtime by 87% monthly.

• Developed ETL pipelines in R to transform, blend, and automate the ingestion of 150 MICS datasets, leading to a 45% increase in the model sample size when integrated with DHS surveys.
• Created impactful data visualizations, effectively presenting statistical and technical information in a clear and concise manner, contributing to transparent understanding of research findings.