Internship Experiences

Accenture 

At Accenture, I started as a Project Manager Intern on a banking digital transformation project, where I coordinated cross-functional squads, tracked MVP milestones, and gained hands-on experience in agile project management within the lending sector. I then transitioned into an AI Software Engineer role, where I built two production-ready chatbots:

• HR Onboarding Assistant - Designed an AI-powered chatbot using Google ADK, Pinecone (for RAG vector search), and Azure OpenAI’s GPT-4o for employee query handling. Developed a modular MCP server with FastMCP for tool orchestration, built a PDF ingestion pipeline using Azure embeddings, and deployed services on Kubernetes with Docker.

• Financial Sentiment Analysis Bot - Engineered a chatbot to analyze public sentiment about companies by integrating Reddit and X (Twitter) APIs, orchestrated via Google ADK for decision-making in the lending sector.
  

This dual role allowed me to understand large-scale project management while also delivering advanced AI solutions from architecture to deployment.

Here is a slideshow of my work on the HR Onboarding Assistant as a AI Software Engineer intern (hover over the photos and click the arrow): 



Tufts Office of University Communications & Marketing
During my sophomore year, I worked as a Data Analytics Intern at the Tufts Office of Communications & Marketing, where I got hands-on experience working with real-world data and turning it into meaningful insights. I was responsible for analyzing two of Tufts’ main websites, Tufts Now and Tufts Solutions, using Google Analytics 4 and Looker Studio to track performance and present findings to stakeholders. One fun moment (and challenge) came when I was asked to manually update a spreadsheet with over 600 faculty names, instead, I built a Python web scraper to automate the entire process, improving both speed and accuracy by over 200%. I also made sitemaps for the Fletcher and Undergraduate Admissions websites to guide long-term planning, and collaborated with my co-intern to create a script that outputs the most common words from any webpage.

This internship showed me just how powerful data analytics can be when it comes to decision-making. The process of turning raw numbers into clear, actionable insights is what helps organizations make smarter, faster choices. Being able to connect the technical side of data with real-world impact is something I now see as essential to any company’s success.


LINK TO FULL DASHBOARD




AI and Robotics Venture: CarivaThis was my first internship. At CARIVA, I was immersed in a fast-paced startup environment, working with cutting-edge technology that was entirely new to me. It made me excited, but scared at the same time. One thing learned early on was that if you ever get stuck, just go on YouTube! You can learn almost everything on there, at your own pace, by yourself. 

During the internship, I organized and synthesized complex medical data using NLP and information retrieval techniques inspired by Stanford’s XCS224U course. I built Python data pipelines with Pandas and spaCy to preprocess and embed medical notes, enabling accurate mapping of doctor instructions to medical terms. Additionally, I developed a medical question-answering chatbot that reached 85% accuracy, leveraging advanced preprocessing and experimenting with models like bioBERT from Hugging Face.