Hi, I'm Sujan Katari, a Computer Science student at the University of Central Florida with a passion for full stack development and generative AI. I'm especially interested in the future of agentic systems and how they can be used to build intelligent, impactful software.
I've had the chance to explore this space through roles at Lockheed Martin and CSX, where I worked on AI research and built LLM-powered agents using tools like LangChain, RAG, and Model Context Protocol (MCP). I've also developed full stack applications from the ground up using technologies like Node.js, React, PostgreSQL, and AWS.
Long term, I want to work at the intersection of AI and software engineering, building smart tools that help people with their lives. If you're working on something interesting or just want to chat, feel free to reach out!
AI system with LangChain, RAG, GPT-4-turbo for 9,000+ repos, docs, and diagrams. Scaled to 100+ engineers.
Automated video labeling, SVM target detection, improved model accuracy and AUC.
ELT pipelines, 10,000+ tickets to Azure SQL, Power BI dashboards for analytics.
Real-time grocery detection with FastAPI, YOLOv5-nano, automated image preprocessing.
Led classes of up to 40 students (ages 3–60+), developed public speaking and communication skills, and earned 4th degree Black Belt (10+ years).
KnightsHub
A full-stack sublease platform for UCF students, featuring verified listings, secure communication, and profile management. Built with Next.js, Node.js, AWS, and PostgreSQL for a scalable, cloud-native experience. Includes Knightro, an advanced AI chatbot that automates workflows and answers campus questions using GPT-4 and RAG.
MindStream turns meeting transcripts into clean, visual mind maps using agentic AI. Built in 12 hours at GemiHacks 2025, it helps neurodivergent and hard-of-hearing users quickly absorb what matters. Powered by Gemini 1.5, FastAPI, and React.
Grocery Detector MVP is a mobile app that uses computer vision to detect and identify grocery items in real time. Built with React Native (Expo) and FastAPI + YOLOv5, it brings object detection to your phone for a seamless, user-friendly experience.
Dermai is a mobile app for skin cancer detection using a 91.2% accurate custom CNN with on-device inference on 10,000+ images. Built for KnightHacks 2024 and winner of 1st Place in the Databricks Challenge, it addresses the critical need for accessible early-detection tools to combat rising skin cancer cases.
Have a question, an idea to share, or just want to connect? I'm here to chat! Whether it's about potential projects, collaborations, or simply saying hello, feel free to reach out. I'll do my best to get back to you as soon as possible!