Nat Fernelius

Hello, I am Nat Fernelius, a Machine Learning/AI engineer based out of New York City.

I am currently a staff member on the EY Tax, Technology, and Transformation Alwin team. I am proficient with major Python machine learning libraries such as scikit-learn and PyTorch and use them regularly for multivariate regression prediction tasks within the domain of tax. Specifically, within tax I often deal with large, imbalanced datasets where precision and compliance to tax law is essential. I also have significant experience developing ETL pipelines between various cloud database solutions for use within ML/AI business applications. Particular database/data engineering experience includes Neo4j and Azure Databricks along with SQL and the Cypher Query Language. As the expansion of Generative AI has fundamentally transformed so much of the working world, I have also been at the forefront of developing generative AI solutions with particular experience with the Semantic Kernel SDK and the Microsoft Copilot ecosystem. Most importantly I am a lifelong learner with a passion for all that the cutting edge of technology has to offer.

Projects

Below you will find an overview of some of my favorite and most meaningful projects that I have created or on which I have contributed.

• Airbnb Data Scraping, Time Series Analysis
• Predicting Unemployment With Sentiment Analysis Methods and Multivariate Regression
• Determinants of Human Development in Sub-Saharan Africa
• This Website

Click HERE for all my published papers including my capstone theses in econometrics and machine learning.

Contact Me

If you would like to get into contact with me about any of my work or future collaborations do not hesitate to reach out below.