Andre Paez Castro
Exploring XR, Intelligent Systems & AI
About Me
Hi, I’m Andre Paez Castro, a Graduate student in Information Science at the University of Pittsburgh, where I currently maintain a 4.0 GPA. I hold a B.S. in Psychology with a minor in Cybersecurity from Western Michigan University, where I was a Dean’s List honoree and inducted into the Tau Sigma National Honor Society.
My expertise lies at the intersection of AI, Intelligent Systems, and Immersive Technologies. I am dedicated to architecting scalable software solutions and exploring how Computer Vision can be leveraged to create more intuitive, high-impact digital experiences.
I build with a versatile stack including Python, Flask, JavaScript, C#, and Unity. Whether developing data-driven backends or interactive XR applications, I am motivated by the challenge of engineering innovative tools that solve complex problems. I’m always eager to connect with like-minded professionals and explore new opportunities in the tech landscape.
Projects
Emotion Detection App
This application performs real-time emotion recognition by analyzing a user's webcam feed. It leverages a lightweight ONNX (Open Neural Network Exchange) model for efficient, client-side inference, paired with a minimal Flask backend to serve the application. The primary goal was to create a responsive tool that demonstrates the power of computer vision models running live on the cloud.
Technical Challenge: Developing for real-world environments highlighted the model's sensitivity to variable lighting conditions. This project served as a deep dive into the importance of data augmentation and the limitations of lightweight models compared to higher-parameter alternatives in non-ideal environments.
This project is deployed on Microsoft Azure and features a full CI/CD pipeline using GitHub Actions for automated builds and deployment.
Teak Custom Woodwork & Interiors
As the sole developer, I architected and deployed a professional digital presence from the ground up to serve as the primary lead-generation platform for a custom furniture business. Developed using Next.js, React, and Tailwind CSS, the site features a responsive showcase of high-end craftsmanship paired with a custom inquiry system.
The application leverages Supabase for backend data management and is deployed on Vercel. By implementing modern SEO practices and automated deployment workflows, the platform successfully converted digital traffic into the company's first-ever online sales leads.
Frontier Hunt (VR Survival Game)
Frontier Hunt is an immersive virtual reality survival experience developed specifically for the Meta Quest 3. Built using Unity and C#, the project focuses on high-fidelity spatial interactions and physics-based logic. It utilizes the Meta XR SDK and OpenXR to implement precise input mapping, including physics-based lantern toggles and environmental interactions.
The project emphasizes performance optimization for mobile VR hardware, leveraging efficient 3D asset management and atmospheric lighting to maintain the consistent high frame rates necessary for player comfort.