For this project we leveraged Natural Language Processing (NLP) and Machine Learning to create a chatbot that understands user input and recommends potential career paths based on their interests,
skills, and experiences. An intuitive user interface (UI) and a conversational flow were designed to make interacting with the chatbot smooth and engaging. By incorporating user testing and feedback, we were able to refine the chatbot’s accuracy and ensure it delivers increasingly relevant career suggestions.
DevRev.ai is a business software company that focuses on bridging the gap between developers and customer support. As a summer intern I was involved in Developing and optimizing LLM models to
enhance developer-to-customer communication at a high-growth startup.
Enhanced data quality by testing and creating datasets to improve LLM performance. Built custom agents on DevRev.ai’s core product, gaining experience in real-world application of LLMs.
This project investigated a novel approach for early detection of Parkinson’s Disease (PD) using DaT SPECT imaging. A Convolutional Neural Network (CNN) model based on the Inception V3 architecture was designed and implemented. Custom layers were incorporated to further enhance the model for the specific task of DaT SPECT image analysis. This approach achieved a significant 23.24% increase in PD
diagnosis accuracy compared to a standard Sequential Model (74.65% vs. 51.41%). This project has the potential for further development and possible implementation in automated PD diagnosis systems.
This app was created for Autistic users who needed a way to keep track of what was in their refridgerators without manually looking through it. While this was a non-coding project, I learned how to go through the design process to create a user friendly app. I worked on this with a group and we spent a lot of time debugging problems and making the app easier for our partner to use. The app doesn’t have a complicated instructions or too many buttons and our project partner felt the app was usable and helpful for her. I used Adalo to create this app. It is available through the qr-code.
This is an initiative that I started to host free virtual Java programming sessions for 6th to 12th graders via San Jose public library platform. I have been an active volunteer at San Jose public library for over a year now and have been trained to host programming sessions in my area of expertise. I host these sessions in Evergreen Village Square Library and Tully Public Library branch.
This box will unlock and lock with the right key fob as well as through your phone. This uses various different compnents such as the bluetooth module, RFID reader, Arduino, servo, and more. I used a combination of hardware engineering as well as software engineering to complete this. I can now solder and code using the arduino libraries.
Hydrate was a buisness pitch that was created to sell a water bottle. This was a water bottle that used sensor technology to monitor water intake as well as regulate your daily needs through an app. While I never created a working product and app, I did create a prototype online as well as a flowchart of how the app would look like. I used figma for the app flowchart and built a website.