João Nuno Valente
- (+351) 910 975 876
- hello@JoaoNunoValente.com
- JoaoNunoValente.com
- Aveiro, Portugal
Ph.D. Candidate building and deploying applications using Python, React and Docker. I develop and maintain projects such as Cantinas.pt, a web platform that aggregates and displays the menus from the University of Aveiro’s canteens, and the Robot Game, a Python arcade game developed with Pygame and deployed via Docker.
I am currently doing by Ph.D at the University of Aveiro, where I apply automation, real-time data processing, and develop workflows to research problems involving 3D laser vibrometry, depth cameras, and robotic-arm systems. I also maintain my Ph.D. website, built with Jekyll, and a minimalist Bio Website built with Pico CSS.
Projects
Cantinas.pt is a free access web platform that aggregates and displays the menus from the University of Aveiro’s canteens. It retrieves real-time data directly from the university’s official API, ensuring accurate and up-to-date information.
The project was designed using a Cloudflare Worker as an intermediary layer for API requests. This setup significantly reduces latency and ensures near-instant load times even during peak usage. The frontend is a React + Vite application.
Robot Game is a simple arcade-style game where the player controls a robot to collect coins, avoid monsters, and restore health using hearts.
The game was developed in Python using Pygame, exported to the web via Pygbag, and containerized with Docker for easy deployment. It’s playable directly in the browser and available as a prebuilt image on Docker Hub.
This project was built as the final project for the University of Helsinki’s Introduction to Programming and Advanced Course in Programming.
Ph.D. Website is a personal academic website showcasing my research progress. The site was built using Jekyll for static site generation, hosted on CloudFlare Pages, and styled for a clean and responsive design.
This platform serves as a central hub for my academic work and is regularly updated to reflect ongoing progress.
A machine learning system that classifies hand-sign gestures for digits 1–10 using CNN and Fully Connected Neural Network models. The work includes end-to-end training, evaluation, dataset cleanup, and visualization of convolutional layers to show how the model processes inputs.
This was the final project of module Introduction to Data Science and Machine Learning . The repository provides Jupyter notebooks for training both models and can be found here.
Bio Website is a minimalist bio page built with Pico CSS and very lightweight. It links to my social profiles, CV, Ph.D. website, and personal projects.
Scholarships / Grants
Ph.D. Research Grant – Single-Laser 3D Vibrometry with Depth Camera Integration
Ongoing Ph.D. research focused on developing a low-cost and flexible 3D vibrometry system using a single scanning laser integrated with a depth camera for positional tracking. The project aims to demonstrate the feasibility of full-field vibration analysis without the need for multi-laser setups.
- Using Python for real-time communication and data acquisition from RGB-D cameras.
- Employing Docker for camera calibration and running point cloud registration methods in isolated environments.
- Developing an autonomous 3D laser vibrometry system using point cloud registration techniques.
- Integrating scanning vibrometry with robotic automation for modal analysis.
The Ph.D. website can be found here.
Research Grant – Development of an Autonomous Sanding System
I developed, designed, and built a prototype of an automated sanding system for Primus Vitória – Azulejos, S.A. The system aimed to maintain constant contact pressure on ceramic tiles during production, minimizing downtime and ensuring consistent surface quality.
- Analyzed and documented the existing manual sanding system.
- Designed a CAD-based prototype integrating pneumatic and load-sensing control.
- Defined and budgeted mechanical and pneumatic components for implementation.