Nov 2020 – May 2021 · 6 mos

• Migrated the data of over 28 websites from Drupal 7 to Drupal 9 with migration tools by creating YAML ETL pipelines and utilizing SQL to understand and manipulate the large database.
• Designed and implemented the ASU's crimeandjusticenews.asu.edu website using up-to-date ASU web standards and front-end design.
• Increased team productivity by 10% through efficient Kanban board management.

Jan 2020 – May 2020 · 4 mos · Arizona, United States

• Classified 50k COVID-19 articles related to vaccines and therapeutics by scraping Google search results to obtain noisy data and training a scientific-text-based Bidirectional Encoder Representations from Transformers (BERT) model called SciBERT.
• Ranked COVID-19 articles for queries relevant to vaccines and therapeutics by utilizing BERT as an embedding generator and finding each article’s Cosine similarity with keywords related to vaccines and therapeutics.
• Identified Randomized Controlled Trials (RCTs) from over 50k highly imbalanced PubMed articles by modifying the BERT architecture and manipulating its inputs along with various NLP techniques using PyTorch and transformers.
• Implemented a visualization to understand BERT’s attention architecture and the effects of modifications made on its inputs.

Jun 2018 – Aug 2018 · 2 mos

• Refactored the core functionalities of LuaRocks commands for - listing, uninstalling, and showing details of packages, searching and installing rocks from the web, opening documentation, linting the rockspec, selecting a rock-tree, etc., to modularize them.
• Programmed a complete Application Programming Interface (API) to provide access to the LuaRocks functionality using Object-Oriented design patterns and used Git extensively for contributing to the main code-base.
• Designed a responsive and interactive web-based GUI using HTML, CSS, Bulma, and Vue.js to give access to the LuaRocks functionality. Interfaced the GUI with the LuaRocks-API in the backend using CGILua.

Aug 2017 – Dec 2017 · 4 mos

• Boosted prospective car sales by up to 19% by predicting prospective car customers using car sales and inquiry data with millions of data points spanning one year using time series features and logistic regression classification
• Supported senior data scientists and analysts on high-impact client projects—leveraging advanced techniques such as one-hot encoding, moving averages, linear regression, spectral clustering, and ARIMA on complex multimodal datasets using Pandas, Scikit-learn, and Alteryx, thereby deepening my ability to transform intricate data into strategic insights.

Jun 2017 – Aug 2017 · 2 mos

• Applied machine learning models like - K Nearest Neighbours (KNN), Support vector machines (SVMs), logistic regression, etc. for classification on various datasets from Rannlab Technologies’s clients by utilizing NumPy, Pandas, and Scikit-learn.
• Spearheaded the creation of a chatbot by implementing a seq2seq model using TensorFlow to showcase it to potential company clients.

Dec 2015 – Jan 2016 · 1 mo

• Marketed about the Umeed NGO to people ranging from school students to senior citizens.
• Assisted more than 20 people to register as a volunteer for the NGO.
• Went to very low-income regions to help and support young children with their studies.