Hello, I'm data scientist with a love for statistics and programming. My primary skillset lies in data wrangling using libraries like Polars or PySpark depending on the task. My main language is Python, however I also do a lot of hobbiest work in R and I'm sure over the course of my career I'll be learning all sorts of weird tools.
| Type | Skill |
|---|---|
| Languages | Python, R, SQL |
| DS Frameworks | PySpark, Polars, Pandas, Numpy, SKLearn, Jupyter, Quarto |
| NLP Frameworks | HF Transformers, SpaCy, NLTK |
| Data Analysis and Visualization | Plotly, GGplot, Matplotlib, PowerBI, Excel |
| Web Frameworks | Flask, Streamlit, R Shiny, RBlogdown |
| Infrastructure | Git, Databricks, Docker, Poetry |
| AWS | Sagemaker, ECR, S3, IAM |
This project seeks to quantify the reduction in pedestrian risk that is expected to result from the installation of traffic calming measures being installed on two intersections on the main road of Granby Colorado. This project heavily leverages Highway Safety Manual (HSM) resources and methodologies.
This project contains:
- An interactive web application containing our reported findings and various interactive tools that empower method exploration.
- All of the notes taken during the project development process.
The Streamlit application serves as both a reporting interface and an interactive analysis tool for the Granby Pedestrian Consulting Project. The purpose of this application is to provide insight into nebulous risk estimates. Giving users the ability to modify variables themselves show the impact they have on risk estimates in real time.
This application is deployed live on streamlit. The link can be found below.
Fire Emblem is a video game franchise with a long history and a whole lot of data. It's a wonderful mix of probability and strategy and, as such, is the perfect place to build a tool with my background. With this project my goal is to create a verstile dashboard giving users access to all of the data they may need at a moments notice and the ability to extract new insights about these games many of us love.
Tools Used:
- Python
- BeautifulSoup
- Polars
- DuckDB
- Plotly
- PowerBI
- Streamlit
I also frequently write about my work on this project on my website! Here is an example of the work that exists outside of the project repository!
I wrote two writeups hosted on my website. The first is a user-friendly introduction to the Huggingface Transformers API. I synthesized a variety of sources to give, what I consider to be, a great primer on fine-tuning an LLM for named entity recognition. That writeup can be found here
The second is a companion piece which seeks to guide new users on AWS how to containerize and deploy an LLM and corresponding Flask app to Sagemaker. This one took ages to figure out myself and I hope it can help others overwhelmed by AWS at the start! Link to this writeup