Vedant Sheel

An autonomous drone system that picks a building or property, maps it in 3D, and flies itself to shoot cinematic footage, no piloting required.

• •Built an autonomous drone system that identifies a user-selected building or property, constructs a 3D representation of the environment, and detects visually significant viewpoints to automatically capture cinematic aerial footage without manual piloting.
• •Implemented optimization-based mission planning using viewpoint scoring, trajectory generation, spline smoothing, and OR-Tools routing to produce safe, physically feasible, and visually optimized flight paths.

Computer Vision · 3D Reconstruction · Path Planning · OR-Tools

A Raspberry Pi robot running QNX OS that understands plain-English objectives and figures out how to execute them itself, with no hand-written computer vision or control logic.

• •Built a modular, MCP-powered (Model Context Protocol) system that lets a Raspberry Pi robot running QNX OS understand and execute natural language objectives through real-time reasoning and dynamic hardware control.
• •Connected a Flask server exposing motion and vision endpoints to an MCP server, letting the LLM plan, act, capture images, and self-correct without traditional computer vision or manual logic.

MCP · QNX OS · Flask · Raspberry Pi · Gemini · Claude

A multi-model system that predicts how pulmonary fibrosis will progress from CT scans, accurate to within ±47 mL of lung capacity, now being validated on real clinical data.

• •Developed a first-of-its-kind multi-model system combining a self-attention CNN with regression to predict pulmonary fibrosis progression, reaching a ±47 mL absolute error rate on lung capacity decline.
• •Integrated a watershed segmentation algorithm for lung volume estimation and Hounsfield Unit normalization for CT preprocessing, outperforming traditional methods.
• •Working with the Firestone Institute for Respiratory Health to validate the model on clinical data, aiming toward peer-reviewed publication.

Self-Attention CNN · Regression · TensorFlow · OpenCV

An autonomous rover that patrols forests looking for wildlife, fires, and waste, cutting manual monitoring time by 70%.

• •Developed an autonomous environmental monitoring rover capable of detecting wildlife, fires, waste, and plant life with 94% classification accuracy using YOLOv8.
• •Integrated high-precision GPS mapping with a MongoDB-backed data pipeline, reducing manual forest monitoring time by 70% compared to traditional field surveys.

YOLOv8 · GPS Mapping · MongoDB

A mobile and desktop app that sorts your trash for you. 97% accurate, in under 250ms.

• •Built a mobile and desktop waste-classification platform using a TensorFlow Lite CNN, achieving 97% accuracy with real-time predictions in under 250 ms.
• •Developed a Flask API and MongoDB backend to manage user data and classification history, training on 50,000+ images to improve waste-sorting behavior by 28%.

TensorFlow Lite · OpenCV · Flask · React Native

A semantic search platform that matches lost items to found ones from a plain-language description, not exact keywords.

• •Built an AI-powered platform using semantic search and vector embeddings to match lost items based on natural-language descriptions, achieving 91% top-3 match accuracy.
• •Integrated the Cohere Vector Database for fast, scalable search across 15,000+ item descriptions, delivering results in under 500 ms.
• •Originally built at Hack the North 2024.

Cohere Vector DB · Embeddings · React · Node.js

A real-time ASL-to-voice translator, 98.4% accurate across 100+ gestures, running under 100ms latency.

• •Developed a real-time ASL translation system achieving 98.4% accuracy across 100+ gestures using CNNs trained on a custom dataset, optimized for sub-100 ms latency.
• •Designed a two-stage recognition pipeline and interactive GUI with integrated text-to-speech for instant voice translation.

CNN · MediaPipe · TensorFlow · Text-to-Speech

A self-orienting solar panel that chases the sun all day and beats a static panel by almost 19% in real-world output.

• •Designed and built a self-orienting solar panel system using dual LDR sensors and servo motors to track sunlight throughout the day, with a real-time PID control system for stable positioning.
• •Achieved an 18.7% improvement in energy output compared to a static panel in real-world testing.

Arduino · PID Control

A model traffic light that reads real traffic distance and cuts unnecessary stops by 30%.

• •Built a model traffic light system using Arduino and ultrasonic sensors to detect vehicles and dynamically adjust light changes based on real-time traffic distance.
• •Wrote logic that reduced unnecessary stops/starts by over 30% compared to traditional fixed-timer lights.

Arduino · Ultrasonic Sensors · C++