Jun 2020 – Jun 2022 · 2 yrs · Boston, Massachusetts, United States

Jul 2018 – Jun 2020 · 1 yr 11 mos · Greater Boston Area

• Data Engineering: Wrote software library for efficiently training deep learning models on 60 hours (200 GB) of log data collected from autonomous vehicles. Engineered a data pipeline for parsing data from vehicle logs, preprocessing data, and storing in MongoDB database. Use of my software was instrumental in releasing CoverNet, a novel deep learning algorithm for predicting trajectories of vehicles with 40% improvement over state of the art. Tools Used: python, PyTorch, MongoDB, PySpark.
• Deep Learning Research: Researched convolutional neural network architectures for predicting future trajectories of vehicles from log data collected from out fleet of autonomous vehicles.
• Tools Used: python, PyTorch
• nuScenes Prediction Challenge: Implemented an open-sourced software library for training and evaluating deep learning vehicle prediction models on the nuScenes dataset. This software library is being used as part of a contest where scientists across the world compete on who can train the most accurate model. See the code here.
• Tools Used: python, PyTorch
• MapManager: Developed a python package for efficiently manipulating map data without consuming all the available RAM. With this package, it is possible to train deep learning models that require map data on datasets that don’t fit in memory.
• Tools Used: python, numpy.
• LaneTarget Estimator: Developed a Random Forest model for predicting the lane a vehicle will take in an intersection. Showed 11% improvement in performance over the existing method.
• Tools Used: pandas, numpy, sklearn.

Jul 2016 – Jul 2018 · 2 yrs · Greater Chicago Area

• Television Station Clusters: Used t-SNE and K-Means algorithm to create television station clusters for use in Nielsen ratings calculations. Use of these clusters improves ratings accuracy by 8%. Created an interactive dashboard in Shiny to visualize clusters and present to clients.
• Tools Used: Pyspark, Python, sklearn, R Shiny.
• Household Demographic Prediction: Helped develop a Recurrent Neural Network to predict household demographics based on cable set-top box data. Researched how to correct model predictions to match census estimates with mixed integer programming. Productionalized model training code to scale to millions of homes.
• Tools Used: Pyspark, Python, Keras, Tensorflow, cvxpy.
• DVD Sales Prediction: Analyzed DVD sales data and developed a random forest model to predict future sales within 15% relative error. Designed an interface with R Shiny to allow stakeholders to make predictions.
• Tools Used: R

Jun 2015 – Aug 2015 · 2 mos · Miami/Fort Lauderdale Area

• Built an Excel dashboard to track food item satisfaction data on a weekly basis. This dashboard can be used to easily flag problematic trends as they arise and make actionable recommendations to field team.
• Redesigned and recoded the existing Excel tool used to rank franchisees in terms of operational performance to make it more helpful to franchisees and easier to use.
• Used Excel, VBA, and R to complete my projects.

Jun 2014 – Aug 2014 · 2 mos · Greater Chicago Area

• Analyzed financial data and ran proprietary models in order to forecast the profitability of proposed business deals. Compiled these findings in an easy-to-read report for upper management to approve or deny the deal.
• Coordinated communications with clients and finalized presentations sent to them.