Austin Berg

CS & Applied Statistics @ UVA

I build data pipelines, train ML models, and analyze data to surface real insights.

About Me

A bit about who I am and what I work with

Hey, I'm Austin — a third-year Computer Science & Applied Statistics student at the University of Virginia. Passionate about machine learning, data engineering, and building systems that turn raw data into real decisions.

Outside of code, I'm chapter president of my fraternity, secretary of Google Developer Groups @ UVA, and an assistant hockey coach for the UVA club team.

3.52
UVA GPA
3
Internships

Experience

AI Software Engineer InternGreekCoreJan 2026 – Present
  • Migrating codebase from Replit to GCP/GCS, establishing cloud hosting and storage infrastructure.
  • Leveraging Claude AI to rapidly implement new features within an agile workflow.
AI/ML InternPelagic AIJun 2025 – Aug 2025
  • Designed a three-table SQLite schema, reducing scenario initialization time by 50%+.
  • Prototyped a Python-based MCP pipeline for automated OSM ingestion and airport graph generation, cutting data prep time by 40%+.
  • Refactored FastAPI endpoints and centralized logging to improve latency and consistency.
Machine Learning InternPelagic AIJun 2024 – Aug 2024
  • Trained regression, classification, and neural-network models on orbital-mechanics data, improving satellite-control accuracy.
  • Delivered production-ready code and visualizations to Jacobs' $16B Aerospace division.

Languages

Python
SQL
R

ML & Data Science

PyTorch
Scikit-Learn
XGBoost
Random Forest
Pandas

Data Engineering

DuckDB
Prefect
FastAPI
SQLite
ETL Pipelines

Visualization

Plotly
Jupyter
Matplotlib

Infrastructure

Docker
GCP / GCS
Git

Projects

Things I've built, either on my own or at my previous internships!

Featured

MLB Pitch Data Engineering & Analytics Pipeline

End-to-end ETL pipeline ingesting, transforming, and validating 140,000+ MLB pitches with Prefect & DuckDB. Engineered 20+ features, deployed a FastAPI service with live ETL triggering, and trained XGBoost & Random Forest classifiers achieving 72% fastball accuracy and 55.5% weighted F1 across 9 pitch classes.

Python
DuckDB
Prefect
FastAPI
XGBoost
Scikit-Learn
Docker

Airport Graph Generation Pipeline

Python-based MCP pipeline for automated OSM ingestion and airport graph generation at Pelagic AI. Reduced data preparation time by 40%+ and enabled rapid updates of airport network models.

Python
FastAPI
SQLite
MCP
OSM

GreekCore Cloud Migration

Migrated GreekCore's full codebase from Replit to GCP/GCS, establishing production-grade cloud hosting and storage. Leveraged Claude AI to accelerate feature delivery within an agile workflow.

GCP
GCS
Claude AI
Cloud Infrastructure

Personal Portfolio

The site you're looking at. Built with Next.js 15, Framer Motion, Tailwind CSS, and shadcn/ui.

Next.js
TypeScript
Framer Motion
Tailwind CSS

Hobbies

What I do when I'm not at the keyboard

🏒

Ice Hockey

Assistant coach for the UVA ACHA II club hockey team. Running practices, calling live strategy adjustments, and developing players on and off the ice.

🧗

Climbing

Bouldering and sport climbing. Always chasing the next problem.

🎸

Guitar

Learning to play. Still figuring out chords but getting there.

🎮

Gaming

Strategy games, roguelikes, and the occasional open-world adventure.

🎧

EDM & House Shows

Catching live sets whenever possible. House, techno, and everything in between.

🍳

Cooking

Experimenting in the kitchen between problem sets.

Contact

Want to say hi? Reach out!