Software developer with a strong attention to detail, adept problem solving abilities, and a desire to make a difference. Experienced with research on modeling human behavior, and currently working as a Software Engineer at Google developing custom audience targeting models at scale.
Overall GPA: 3.81 (Out of 4.00)
Cum Laude with Highest Distinction
Activities: Golden Key International Honour Society, Human Computer Interaction Lab (ROC-HCI), Computer Science Undergraduate Council (CSUG), Intramural Soccer and Ultimate Frisbee
Modeling and development of custom audience solutions for unified cross-platform (YouTube, Google Search, Google Display Network, etc.) audience targeting. Based in the global headquarters in Mountain View, California.
Apply machine learning techniques to perform automated lie detection from audio and video at the University of Rochester HCI Lab (ROC-HCI). Use hidden Markov models and clustering algorithms to recognize patterns in human conversation. Deploy code on BlueHive supercomputing cluster to train and test models on massive dataset.
Assist graduate students and professors in troubleshooting technical problems and configuring devices. Automated printer configuration process for students by developing a one-click application in AppleScript.
Mentor project teams in designing and building products to meet a specific consumer need. Grade assignments and hold weekly office hours for Human Computer Interaction course.
HTML & CSS
Leverages the Spotify Web API and scikit-learn to analyze users’ top 100 tracks to identify their music taste. Analyzes multiple users’ playlists to create a shared playlist suited for a group activity, such as studying. Uses machine learning to generate a playlist of songs tailored to an individual user’s music preferences. Read more about it here or check out the code on GitHub here.
Finds shortest paths in a graph from a source node using a parallelized delta-stepping algorithm in Java. Achieved nearly three-hundred percent speedup compared to the sequential version of the algorithm.
Website and iOS app to track university shuttles in real time and find optimal routes using a graph algorithm. Developed a custom API with Node.js utilizing Google Maps API and Transloc API for shuttle information. Worked with one peer on engineering side, while other teammates focused on user research and evaluation. Check out the code for the web app on GitHub here.
Compares inference algorithms on probabilistic graph models in Java for Artificial Intelligence course. Individually created an exact calculator and several approximation algorithms to comply with larger datasets.
T. Sen, K. Hasan, M. Tran, M. Levin, Y. Yang, and M. E. Hoque, Say CHEESE: Common Human Emotional Expression Set Encoder and its Application to Analyze Deceptive Communication, IEEE International Conference on Automatic Face and Gesture Recognition, Xi'an, China, May 2018.
More in the works, check back soon!
Academically, I am primarily interested in machine learning and artificial intelligence. I love working with big data to find an interesting pattern, and figuring out how to apply these insights to a real world problem, ideally to improve the lives of those around me.
Outside of computer science, I am an avid guitarist and bassist, and have spent many summers working as a music instructor teaching guitar, bass, and drums, as well as various aspects of musicianship such as songwriting and stage presence.
I'm also a soccer fan and enjoy playing ultimate frisbee in my spare time. I have a passion for coffee and am a loyal dog-person (my apologies to cat-people).