• Created a novel and efficient method for accessing information on Texas bill turnout • Collaborated in developing an NLP algorithm that will predict whether bills will pass in the Texas Legislature • Trained and implemented an LSTM classification model with a 90% accuracy rate, using the bag of words approach, to predict bill pass or fail percentages • Optimized model scoring pipeline through quantization, data down-sampling, and algorithm refactoring, resulting in a significant increase in accuracy percentages