Software Engineer & AI Enthusiast
I strive to be an ethical computer scientist simultaneously building and researching like no other. I'm a sophomore at Georgia Tech studying Computer Science with a focus in Intelligence and Information Internetworks.
I'm a sophomore Computer Science student at Georgia Tech, specializing in Intelligence and Information Internetworks. My interests lie at the intersection of software engineering and artificial intelligence — particularly in building scalable systems and researching ethical AI applications.
As the founder and CTO of Univo and a former SWE intern at Hachi, I've built and shipped full-stack applications using modern frameworks like React, Node.js, Firebase, and Redis. I’ve also contributed to open-source community tools and am currently engaged in AI research on model jailbreak evaluation and automated algorithm design.
Outside of code, I explore new AI papers, play soccer, practice percussion, and design platforms to empower the communities I care about.
Years Experience
Projects Completed
Technologies Mastered
Georgia Tech
Georgia Tech
Georgia Tech
Georgia Tech
Georgia Tech
Georgia Tech
Georgia Tech
Georgia Tech
Georgia Tech
Georgia Tech
Georgia Tech
Remote-sensing segmentation pipeline leveraging Sentinel-2 to replace clouded Landsat scenes, histogram normalization for contrast, and tuned training (AdamW, Dice/Tversky losses) for higher accuracy and faster convergence.
QR code ordering system enabling seamless table-side food ordering and payments across hospitality venues. Successfully secured $500,000 in investment funding.
Framework to evolve hybrid algorithms from human baselines and study them for new insights.
Live meeting tool with real-time transcript sync, agendas, and analytics with low-latency caching.
Researching systematic jailbreaks to induce biased outputs; building an evaluation harness.
Personalized fashion hub for trend exploration and curated styles; rebuilt backend for real-time sync.
Flask-based funeral services app with SQLite backend, and OpenEMR integrated app made for Masjid Al-Farooq.
CNN trained on chest X-rays; 92% test accuracy with preprocessing to cut false negatives by ~15%.
Social platform linked with workouts; uses Gemini for personalized routine recommendations.
QR code ordering system enabling seamless table-side food ordering and payments across hospitality venues. Successfully secured $500,000 in investment funding.
Live meeting Zoom App with in-meeting analytics, OAuth-protected routes, and low-latency rollups via Redis caching.
Flask-based funeral services app with SQLite backend, deployed and currently in use.
Social platform linked with workouts; uses Gemini for personalized routine recommendations.
Utilized Sentinel-2 to cut 170 clouded/low-quality samples, applied histogram normalization for multispectral contrast, and tuned AdamW with Dice/Tversky losses—improving segmentation accuracy by ~13% and speeding convergence ~21%.
Evolving hybrid algorithms from human baselines to inspire new methods.
Evaluation suite for bias-inducing jailbreak prompts; WIP.
92% test accuracy; preprocessing reduced false negatives ~15%.
I'm always interested in new opportunities, collaborations, and interesting projects. Feel free to reach out if you'd like to work together or just want to say hello!
