About Me

My interest in data science started in undergrad when I took a business analytics course during my spring semester of senior year. Of course, at that point, it was too late to change my major to statistics or computer science, so I decided to get my master’s degree. While working at Xcel, I took classes in programming and advanced math to prepare for grad school. I ultimately left my job at Xcel to focus on learning data science at Berkeley full time.

Since graduating from Berkeley, I have focused my career on the development side of data science. I still love cleaning data and writing models, but my skillset and interests will best match a position where I develop backend software and work with data scientists to integrate models into the data pipeline, and ensure that the whole system is reliable, scalable, and easily monitored. I’m deeply passionate about solving the puzzle of making the whole process flow as efficiently as possible with tools like Airflow and AWS.

As a decision scientist at Ibotta, my main project involved automation of data analysis using Airflow. I then pursued an opportunity to deepen my knowledge of software engineering at Amazon, where I used AWS tools such as Lambda, CloudFormation, and StepFunctions to construct a machine learning platform for AmazonGo. I have learned a lot of valuable skills that I hope to take with me into my next role.

If you catch me outside of work, I’m an adventurer who enjoys rock climbing, skiing, and mountaineering. I’m also currently converting a Dodge ProMaster into a campervan! I have never done any sort of “handy” work like this before, but it has been so rewarding to come up with an idea, watch YouTube videos and read blogs, then implement the vision myself.

At this point, I am not looking for just a job to pay the bills; I’m looking for a career that I can grow in for a long time. I am looking for a role that provides challenging and meaningful work, while also providing good work-life balance. I live in Boulder, CO; my perfect workday would be an early morning start, sipping tea and writing code, being quite physically sedentary but mentally engaged all day, then finishing my work and heading to the crag for a couple of hours of climbing and fresh air before it gets dark.

Portfolio

Machine Learning Projects (in Python)

Mountain Project Sentiment Analysis using Transfer Learning and DistilBERT

This end-to-end project uses deep learning for sentimant analysis of natural language. The goal is to label the sentiment of Mountain Project forums to determine what gear climbers think to be the best. The writeup follows the process from webscraping, to data cleaning, to model selection, and final analysis. This was a solo project.

---

Audio Super-Resolution - Thesis Project

This project uses GANs and Autoencoders to upsample low quality audio in real time. This was a team project.

---

Car Damage Detection Part 1 (Image Classification)

This section uses traditional machine learning (KNN, Naive Bayes) and CNNs (with and without dropout) to classify images of cars as damaged or whole. This was a solo project.

Car Damage Detection Part 2 (Identify damage location)

This section uses Mask R-CNN to identify the location of the damage on the car. This was a team project.

---