Academic Projects

Project Overview Developed within a team of 6 students, OccuPI is a fully integrated IoT system that tracks room capacity and occupancy, designed for efficient facility management. Key Implementation Details Edge Detection: Programmed Raspberry Pi devices to analyze local camera feeds using YOLO computer vision networks, detecting person coordinates at the edge. Secure Networking: Configured secure MQTT (MQTTS) and HTTPS network protocols to transmit room count telemetry to a central Debian system. Central Storage & Analysis: Engineered a backend to receive inputs, saving them into a PostgreSQL database. Created a responsive web dashboard for administrators to monitor rooms and view historical data. 🔒 Note: This is a collaborative academic project, the source repository is kept private.

Project Overview An academic project built with a team of 4 students to deploy private, offline artificial intelligence. The objective was to design a system that automates downloading and running large language models locally with hardware acceleration. Key Implementation Details Containerized Orchestration: Developed an integrated Docker Compose environment that automates installing and running Llama.cpp libraries. Hardware Pass-through: Configured NVIDIA Container Toolkit integration to leverage CUDA on host GPUs, accelerating inference speeds. Quantization Pipeline: Automated model quantization scripts to output GGUF formatted models, reducing local RAM overhead while maintaining response quality. 🔗 GitHub Repository

Project Overview Developed in collaboration with an academic team of 4 classmates, this project provides a full-stack dashboard built to collect, structure, and visualize live data streamed from remote IoT sensors. Key Implementation Details Scraping Daemon: Programmed a persistent Python worker background thread to monitor and parse incoming MQTT broker payloads. Data Indexing: Configured a Django backend mapped to a MariaDB database to structure, save, and retrieve telemetry data points. Interactive UI: Engineered a clean, interactive user dashboard using Chart.js to render real-time charts, featuring date filters and a raw CSV data exporter. 🔗 GitHub Repository