May 2023 - April 2025IRIS NITK
Product Lead
I oversaw the complete product lifecycle of a student-built, comprehensive ERP system designed to digitize over 55 academic and administrative processes, positively impacting more than 24,000 users. My responsibilities included preparing detailed Software Requirement Specification (SRS) documents, engaging with key stakeholders, developing intuitive wireframes, and conducting functional testing for individual modules. I led a team of 10 Product Managers and collaborated cross-functionally to ensure timely delivery and continuous enhancements across both web and mobile platforms. Additionally, I spearheaded the digitization of the college’s Accounts and Establishment section as part of a consultancy project, driving institutional transformation through technology.
July 2024 - December 2024O9 Solutions Inc.
Software Developer Intern
I designed and developed a chatbot agent to streamline dataset manipulation within the O9 platform, significantly enhancing productivity for both internal teams and clients. This project involved close collaboration across multiple sub-teams to address diverse use cases, ensuring alignment with the objectives and strategic goals of the PMM Team. Leveraging the HuggingFace library, I implemented BERT transformer models for natural language understanding, and used Streamlit to build and deploy the application, delivering an intuitive and efficient user interface.
April 2024 - May 2024Inovaare Corporation
Intern
I contributed to the Usher AI product, a generative AI platform that leverages Large Language Models (LLMs) and Deep Learning techniques to automate complex NLP tasks. As part of this work, I developed a solution to summarize intricate medical jargon found in MD Decision Rationale documents using GPT-3, automating a traditionally manual and time-intensive process. The summaries were evaluated using standard NLP metrics such as ROUGE and BLEU scores to ensure quality and accuracy. Additionally, I built a BERT-based text classification model for A&G Case Categorization, trained on labeled incident data to effectively determine whether case descriptions should be classified as Appeals or Grievances.
April 2024 - June 2024NITK School of Management
Research Intern
I conducted a comprehensive study on the underpricing of IPO shares by analyzing media sentiment across 300+ companies and their associated news articles. The project began with the implementation of standard lexicon-based sentiment analysis using scoring metrics from VADER, AFINN, and TextBlob to evaluate the impact of pre-IPO sentiment. To enhance accuracy and ensure relevance in financial contexts, I applied FinBERT, a BERT-based language model fine-tuned on financial text. Finally, I employed decision tree models to uncover and interpret relationships between media sentiment and IPO underpricing outcomes, providing valuable insights into how public perception influences market behavior.