Trained to be a Data Scientist but a software builder at heart.
Live stock sentiment analysis web app with a distributed financial news processing pipeline made with Apache Airflow, Apache Spark, and a PyTorch transformers model. The application sits on a highly scalable AWS backend designed primarily around AWS Lambda, Elastic MapReduce, and DynamoDB.
Launched a Data Science interview training website as the DevOps Engineer in charge of maintaining the AWS backend, deploying the internal NLP model API, and reviewing the frontend components.
Ongoing investigation of whether it is possible to train a Reinforcement Learning agent to play a complicated parkour obstacle map in Minecraft, starting from toy obstacles to popular courses played by humans.
1. Built a webcrawler and inverted index search engine with Jetty Servlets in Java.
2. Re-implemented a UNIX shell in C.
3. Wrote from scratch in python: Naive Bayes Classifiers, Decision Trees, Random Forest, Regularized Regression, ADABoost Trees, Gradient Boosting Trees, K-Nearest Neighbors
Send me an email at sean@seantey.io!
