Jun 2024 – Aug 2024 · 2 mos · Mumbai, Maharashtra, India · Hybrid

• Designed and tested over 100 alpha factors using the Fama-French 5 model and TA-Lib, analyzing fundamental and technical variables across 2,218 companies. Enforced factor investing principles to systematically identify and leverage key factors driving market returns with dynamic rebalancing to optimize risk-adjusted returns.
• Deployed a suite of machine learning models including Regression, Random Forest and XGBoost to accurately forecast 1-month forward stock returns. Improved model precision using RFE by emphasizing diversification across factors with a focus on top-performing factors.
• Utilized expanding window and block cross-validation frameworks to rigorously test and validate models within a time series framework, resulting in a 12% enhancement in model stability in predictive performance over time.

Dec 2022 – Mar 2023 · 3 mos

• Developed and launched an education portal, consolidating educational resources from diverse fields; facilitated access for 10,000+ students and professionals, fostering continuous learning and growth.
• Utilized Angular to design the user interface, ensuring a seamless and intuitive experience for the users.
• Integrated a MongoDB database to handle a data volume of over 1,00,000 educational resources efficiently, improved user management functionalities, developed API, and handled server-side processing achieving an uptime rate of 98% and delivering a responsive user experience.

Apr 2021 – Nov 2021 · 7 mos

• Previously known as Madhuriya Art House.
• Worked as an event manager and marketing intern for an event called Aprajita which had artists from around the globe performing on Western as well as Indian music.
• Managed social media accounts by documenting social media strategy, content creation, creating calendar, managing and monitoring engagement and activities as well as worked on social media campaigns among others.

Apr 2021 – May 2021 · 1 mo

• Engineered a trained machine learning model capable of classifying movie reviews as positive or negative to analyze sentiment in unseen movie review.
• Compared various machine learning algorithms, including Naive Bayes, Support Vector Machines (SVM), and Random Forest to determine which one achieved the highest performance for sentiment analysis on the IMDB dataset.
• Achieved a sentiment analysis accuracy rate of 90% with the SVM, outperforming competing models by a significant margin.

Jul 2020 – Aug 2020 · 1 mo

• Worked on market research where we analyzed the potential markets and requirements to start and thrive in countries in Africa and Middle East, thus pitching a marketing strategy. Also worked on content creation for various social media.