Niki Vasan | Data Scientist I @ Truist Bank

Niki Vasan

Data Scientist I @ Truist Bank

Based in Atlanta, GA

About

I am a motivated new grad (Emory University c/o 2023) who is passionate about using data to further understand and explore the world around us. I have experience in all stages of the data science life cycle, with a focus in machine learning and AI.

Skills

Projects

CS470 Spotify Data Clustering Project

This project uses pre-scraped Spotify song data and unsupervised clustering models to understand how quantitative measures of musicality (e.g. tempo, danceability, instrumentalness) correlate to the genre profile of a song or group of songs.

CS334 Computer Vision Project

This project seeks to build upon previous applications of deep learning models in healthcare by evaluating potential biases in a convolutional neural network (CNN) trained to classify comorbidities in chest X-rays using the MIMIC-IV and MIMIC-CXR datasets.

QTM340 Spotify Sentiment Analysis Project

This project seeks to analyze song lyrics of the top albums from Billboard's Top 200 of 2020 list for both male and female artists. We ran a sentiment analysis using NLTK's VADER sentiment analyzer to determine if top selling albums by female artists differ in sentiment from male artists.

QTM30 Cloud Computing Project

Our project tests the interaction between two AWS Machine Learning APIs: Amazon Polly and Amazon Transcribe. We look test for potential hidden variables that could impact the efficacy of either AP, including words of varied origin or spelling, slang or dialects.

2020 Policing Lab Research Project

As an RA for an interdisciplinary lab at Emory University working with policing data, I focused on understanding the spatial concentration of police use-of-force incidents in relationship to minority concentrations in urban counties.

Facebook Data Challenge

As a finalist for the 2021 Facebook Data Challenge, we were asked to consider a hypothetical world in which Facebook is trying to successfully enter the video entertainment streaming industry. This project contains my final pitch and content recommendations for this prompt.

Experience

Jun 2020 - Aug 2020

Summer Intern @ Goodwin Simon Strategic Research

• Qualitatively categorized and analyzed data on over 40 participants as part of an intern cohort for a study by the Bill and Melinda Gates Foundation on youth occupational identity • Gained exposure to different approaches to gauging public opinion such as dial tests, media audits and surveys

May 2021-Aug 2021

Product Analyst Intern @ Mailchimp

• Generate SQL queries to identify and automatically refresh target user volume for new initiative to increase conversion rates across Web/Commerce product suite • Design dynamic visualizations in Looker for Mailchimp’s internal Product and Commerce dashboards • Assist in strategy and implementation of 30 Day report to evaluate early performance of new e-commerce launch

May 2022-Aug 2022

Data Science Intern @ Meta

• Analyze user variation in Facebook Avatar category and content preferences, implement a heuristic ranking A/B experiment to test personalization of content discovery and create a dashboard to visualize ranking output • Build logistic regression and random forest models to systematically identify most significant drivers of adoption • Design framework for future model

Jun 2023-Aug 2023

Data Science Accelerator Intern @ Meta

• Built NLP-based recommender system for 2023 DSAI Hackathon that uses tf-idf vectorization and cosine similarity to suggest investment products for clients • Iteratively design and implement PySpark script that leverages financial transaction data to identify secured lending needs for clients

2023-Present

Data Scientist I @ Truist Bank

• Conduct research into integrating ML solutions (Random Forest, Lasso Regularization, XGBoost etc.) to existing model development workflow for time-series loan forecasting models used in CCAR stress-testing reports • Design and execute end-to-end model development pipeline, assess changes in performance compared to current approach • Migrate dataset from SAS servers to cloud-based data warehouse

Education

2019-2023

Emory University - B.S. Quantitative Sciences

I majored in Quantitative Sciences with a concentration in Sociology and minored in Computer Science on the informatics track. I had a cumulative GPA of 3.94 and a major GPA of 3.96. Some of my relevant coursework includes Machine Learning, Cloud Computing, Database Systems, Probability and Statistics, Linear Algebra and more. At Emory, I was involved in the following extracurricular activities and societies: Office of Spiritual and Religious Life (OSRL) Student Worker, Quantitative Theory and Methods (QTM) Ambassador, Emory Career Center Ambassador, Undergraduate Research Assistant, Phi Eta Sigma Honor Society.