ML Engineer & Systems Thinker

I build production-grade machine learning systems, with a focus on scalable NLP pipelines and real-world deployment. My work ranges from analyzing climate anxiety across millions of social media posts to designing efficient forecasting systems for atmospheric data.

I approach machine learning as an infrastructure problem, focusing on systems that are scalable, reliable, and maintainable. I favor simple, effective models over unnecessary complexity, and rely on rigorous validation and error analysis to ensure performance. Clear documentation is a priority so that systems remain understandable for both technical and non-technical stakeholders.

Resume GitHub LinkedIn

Recent Projects

Featured work with live demos and technical breakdowns

NLP Pipeline for Climate Anxiety Analysis

Analyzed 270GB text data from 21M social media posts using transformers. Cut processing time from 2 weeks to 15 hours.

Customer Feedback Classification System

Scalable NLP pipeline using PySpark and HuggingFace transformers for text classification. Deployed on Databricks with MLflow tracking.

See all projects →

Background

B.S. in Mathematics (Purdue) with Data Science focus and Atmospheric Science minor.

Experience across NLP, time series forecasting, and optimization, including:

• Large-scale NLP pipelines for mental health analysis
• Climate forecasting using ensemble methods
• Optimization-based classifiers (LP, SVM)
• Data workflows for atmospheric datasets

Tech Stack

• Languages: Python, SQL
• ML: PyTorch, Transformers, Scikit-learn, TensorFlow
• Platforms: Azure ML, Databricks, PostgreSQL
• Data: PySpark, Pandas, SQL

What I’m Interested In

• Efficient inference and model compression for real-world deployment
• Production ML systems that scale beyond prototypes
• Bridging research ideas into reliable, maintainable systems
• Teaching practical machine learning (not just theory)