I'm James, a Computer Science student at Partido State University who is fueled by curiosity, coffee, and the drive to build useful technology.
I work as an Android + Computer Vision / Edge AI Developer, focused on building practical apps powered by lightweight AI models. My work combines native Android development, OCR, educational video enhancement, super-resolution, visual classification, TensorFlow Lite deployment, and offline mobile-first systems.
I have always been fascinated by how things work under the hood. That curiosity led me to become the President of SPECS, where I get to share my passion for tech with others, and it drives me every day to build software that actually solves real-world problems.
Whether I'm up late figuring out a complex bug in Kotlin, fine-tuning a lightweight model, testing OCR readability, sketching a new UI, or learning a completely new framework just for fun, I'm always looking for the next challenge. I don't just write code; I try to build tools that help people—like emergency apps, educational systems, offline AI tools, and student-focused platforms.
- My Current Obsession: Native Android Development, Computer Vision, Edge AI, scalable Backends, and Embedded Systems.
- My Goal: To bridge the gap between complex theory and practical, user-friendly applications.
- My Vibe: Always learning, always building.
| Role | Experience | What I Focus On |
|---|---|---|
| Android Developer | 3 Years | Building native Android applications using Kotlin, Java, Jetpack Compose, XML, Firebase, Room DB, Android SDK, and offline-first app architecture. |
| Web Developer | 2 Years | Creating responsive web systems, dashboards, and student-focused platforms using JavaScript, TypeScript, React, Bootstrap, Firebase, Appwrite, and Netlify. |
| Computer Vision / Edge AI Developer | 5 Months | Training and fine-tuning lightweight AI models for super-resolution, OCR, visual classification, readability improvement, and Android-ready TensorFlow Lite deployment. |
| IoT Developer | 2 Months | Prototyping embedded and connected systems using Arduino, ESP32, sensors, and hardware-software integration. |
Languages
Mobile & Frameworks
AI, Computer Vision & Edge Deployment
Backend & Tools
Embedded & IoT
| Project | What is it? | The Tech |
|---|---|---|
| EduScale | A research monorepo for improving the readability of low-resolution educational videos on budget Android devices. I fine-tuned lightweight SPAN super-resolution models, evaluated OCR readability, and packaged optimized TensorFlow Lite models for offline Android enhancement. | Python, PyTorch, SPAN, TensorFlow Lite, Kotlin, Android |
| Illegible Handwriting OCR | An OCR and Android project for recognizing difficult handwriting. I built the Python training workflow, dataset structure, evaluation pipeline, benchmark tracking, and TensorFlow Lite export flow for Android deployment. | Python, TensorFlow, OpenCV, TFLite, Android |
| SafeAI Buddy | An Android screen-safety app backed by a local TensorFlow Lite visual classifier. It runs detection on-device using screen frames, evaluates visual risk locally, and can show a blocking overlay based on configured thresholds. | Kotlin, Jetpack Compose, MediaProjection, TensorFlow Lite, Python |
| CREST | A research repository I built to help students access papers easily. It was a fun challenge migrating legacy code to modern Kotlin! | Kotlin, Jetpack Compose, Firebase |
| SalbaBida | An emergency app designed to keep people safe. It pulls real-time weather and map data to help during disasters. | Java, XML, Firebase |
| HGDG-GAD | A compliance tool for DepEd projects. I focused heavily on making sure the data stays secure and accessible 24/7. | JS, Vite, Firebase, Netlify |
| SPECS System | An admin tool for my own student org! It visualizes data and helps us manage tasks way more efficiently. | Appwrite, Bootstrap, Chart.js |
| Echo-Nnect | A sleek note-taking app. I wanted to build something that worked perfectly offline for rapid thought capturing. | Java, Room DB, Android SDK |
