AI/ML Developer
I'm an AI/ML Developer with hands-on experience in hackathons and independent projects, specializing in real-world system development with AI integration. I've designed a keyboard-free Tetris game using a custom PyTorch CNN, achieving 96-98% validation accuracy and less than 5 ms inference time for classifying hand-drawn tetromino shapes from live webcam input. I'm now building toward a Full Stack Developer role, leveraging my skills in frontend and backend development to create scalable, intelligent applications.
Designed a keyboard-free Tetris game where a custom PyTorch CNN classifies hand-drawn tetromino shapes from live webcam input using Python, PyTorch, MediaPipe, OpenCV, Streamlit, and CUDA.
Built a 4-agent AI pipeline for thyroid risk assessment using Python, scikit-learn, RAG, and Streamlit, developed at AgentX national hackathon securing 3rd place.
Built a multimodal deep learning system for natural language caption generation using Python, PyTorch, HuggingFace (ViT + GPT-2), and Streamlit.
Engineered a real-time pose estimation system tracking 33 body keypoints using Python, MediaPipe, OpenCV, and NumPy.
Predicted cyclone intensity categories and computed optimal evacuation routes using Python, scikit-learn, Graph Algorithms, and Meteorological APIs.
2024 – Present
Secured 3rd Place at AgentX Hackathon with ThyroAssist, built and shipped 6+ end-to-end AI/ML projects across computer vision, NLP, multi-agent systems, and full-stack web applications. Developed SafeRoute AI, a disaster risk prediction and evacuation routing system using XGBoost, NetworkX, FastAPI, and Google Maps JS API.
B.Tech CSE (AI & ML) • 2023 – 2027
8.58 / 10
Intermediate (11th–12th) • 2021 – 2023
Class X • 2021
Secured 3rd place with ThyroAssist, a multi-agent AI clinical decision support system.
Interested in collaboration or just want to say hello? Feel free to reach out!