PROJECTS
Collaborate on cutting-edge AI projects, gain hands-on experience, and contribute to innovative solutions at Georgia Tech.
GEORGIA TECH RAG CHATBOT
With NVIDIA, we built a scalable AI chatbot using NVIDIA Nim models and a vector database for efficient information retrieval.
COLLABORATIVE MULTI-AGENT SLAM
The envisioned final product is an efficient and scalable multi-agent SLAM system which will allow for multiple autonomous robots to collaboratively map and navigate unknown environments with high accuracy, efficiency, and scalability. While this has been explored previously, the CAM-SLAM project aims to enhance these systems by integrating reinforcement learning, deep learning techniques and optimized communication protocols.
INTELLIGENT MALWARE DETECTION
We aim to perform static analysis on software binaries to determine if they are malicious or benign. We will construct Control Flow Graphs (CFG) of the execution paths of suspicious binaries and use architectures such as Graph Neural Networks and Convolutional Neural Networks to predict whether a malware is benign or malicious based on the constructed CFG. For these architectures, we will then perform inference speedups and aim to deliver SOTA models for inference speed and classification accuracy.
