Neha Ramakrishnan

Results-driven Data Scientist with experience at Warner Bros Discovery, adept in Python and SQL. Successfully implemented machine learning algorithms and designed impactful data visualizations, enhancing decision-making processes. Proven ability to collaborate effectively with stakeholders, fostering data-driven insights and optimizing customer segmentation strategies. Strong analytical skills complemented by a commitment to excellence.

Neural Networks, UC Berkeley, Academic Project (Intro to AI), 12/22, Developed and optimized a Python neural network mini-library with custom node types for advanced matrix operations. Showcased expertise in neural network design and implementation, by effectively designing network architectures, fine-tuning hyperparameters, and meeting performance requirements across diverse machine learning challenges. Data for Good Hackathon, JPMC, 10/22 Greater Good in Education, Student Researcher (Data Science Team), 08/21 - 05/22, Conducted extensive EDA and Inferential Analysis on Google Analytics survey data, extracting pivotal insights. Formulated intricate SQL queries and harnessed Pandas tables to distill complex user behavior patterns, transforming raw data into actionable knowledge. Collaborated with cross-functional experts to deliver compelling findings, amplifying user engagement and website performance through data-driven decisions. Air Quality Index (AQI) Prediction, UC Berkeley, Academic Project (Data Science), 11/21, Designed, implemented, and A/B tested multiple regression and logistic regression models to predict AQI for California counties based on the wind conditions of the county and pollutants in its surrounding areas. Utilized Pandas, Seaborn, and scikit-learn for data cleaning, visualization, exploratory data analysis, feature engineering, machine learning, and modeling (both supervised and unsupervised). Movie Classification, UC Berkeley, Academic Project (Data Science), 12/20, Created a k-nearest-neighbor classifier using Jupyter Notebook and NumPy to predict the genre of movies (94% accuracy) from their script dialogue.

Emma Bowen Fellow, 2022- Present