Hi, I'm Dev Yadav, a third-year B.Tech student at VIT-AP University specializing in Computer Science and Engineering with a focus on Artificial Intelligence and Machine Learning. I am deeply passionate about AI/ML, which forms the core of my technical portfolio. I enjoy building end-to-end applications and deep learning pipelines using technologies like Python, PyTorch, OpenCV, React.js, and AWS. Some of my recent projects include LunarVision (a PyTorch U-Net pipeline for lunar terrain segmentation), a real-time facial landmark Fatigue Detection System, and a prototype dashboard for the Smart India Hackathon 2026 (Space Technology). A major recent milestone for me was winning 2nd prize in the GDG AI Prompts War 2026. Beyond technical development, I strongly believe in community building and leadership. I currently serve as the President of the Appdu Gujarat cultural club at my university, where I recently led the organization of large-scale collaborative events like EKTA Garba Night. Previously, I had the honor of leading the Gujarat state rally as the student state coordinator during our annual cultural fest, VITopia. I am always eager to connect with fellow developers, collaborate on innovative AI solutions, and continue growing in the tech community!
LunarVision — Autonomous Lunar Mining Command Center
Architected an AI-powered software prototype for autonomous lunar rovers to mitigate the 2.5-second Earth-Moon communication lag. Developed high-performance deep learning inference pipelines utilizing computer vision to identify, classify, and sort lunar minerals (Ilmenite, Anorthite, Regolith) in real-time. Built a web-based Command Center dashboard delivering live telemetry streams, dynamic bounding box overlays, and automated sorting decisions on video feeds.
AI-Based Online Fatigue Detection System
Engineered a real-time computer vision pipeline utilizing OpenCV and MediaPipe to extract facial landmarks and eye/yawn aspect ratios from live webcam feeds. Trained and evaluated machine learning classifiers using Scikit-learn to classify driver/student fatigue levels with high precision under varying lighting conditions. Optimized frame-processing loops to maintain smooth 30+ FPS performance for real-time workplace and online learning monitoring.
Hostel Management System
Architected a full-stack administrative platform featuring student registration, room allocation routines, and automated fee processing. Designed and normalized a relational database using PostgreSQL to handle concurrent CRUD operations and ensure data integrity.
AI Battlefield Surveillance System (In Progress)
Developing an autonomous drone surveillance module leveraging deep learning framework architectures for real-time object detection and threat recognition. Implementing an automated alert generation system to notify operators of unauthorized target incursions in low-visibility environments.
Intermediate
Vedant Internation Public School
Attendee · Oct 24, 2026