Full-Stack & AI Developer · Philippines

hello world!

I'm Stanley, and I dabble in full-stack development, ai & automations!

View projectsGet in touch

About Me

John Stanley Altonaga portrait

John Stanley Altonaga

Full-Stack & AI Developer

Summary

Based in the Philippines, I'm a computer science graduate of Silliman University, building at the intersection of web development, automation, and AI.

My focus today is on generative and agentic AI, and on designing multi-pipeline systems where data, models, and automation click together like puzzle pieces.

I'm passionate about turning ideas into real-world solutions, earning recognition in regional and provincial hackathons.

Outside of code, I'm drawn to neuroscience and the future of brain-computer interfaces. Off the clock, I'm more or less rotting in video games

Tech Stack

Web & Backend

  • Next.js (React)Core
  • TypeScript
  • Tailwind CSS
  • Node.js (Express)
  • FastAPI / Flask
  • MongoDB
  • Firebase
  • Supabase

AI & Automation

  • PythonCore
  • OpenAI
  • LangChain
  • n8n
  • RAG
  • PyTorch

Cloud & Tools

  • AWS (EC2 / S3 / Lambda / API Gateway)
  • Docker
  • Git
  • Google Apps Script
  • WordPress
  • C/C++, Java, PHP

Experience

AI Automation Engineer (Project-Based) @ Mun. of Sta. Rosa

Nov 2025 - Present · Remote

  • Architected LLM-driven automation pipelines with n8n, OpenAI, and vector embeddings that improved throughput by 60% and reduced latency from 2 minutes to 2 seconds.
  • Reduced token usage, inference latency, and operational cost by redesigning workflows around embedding-based reuse while maintaining high classification accuracy.

Let's Connect

Awards & Recognition

  • ChampionCan You HackIT 2025 Regional Hackathon by IBPAPJune 2025
  • Recognition by Industry LeadersInternational IT-BPM Summit 2025 by IBPAPSeptember 2025
  • 1st Runner-UpAI.deas for Impact by DICTJune 2024

Projects Highlight

Tanglaw facial recognition dashboard

Tanglaw

2026

Initially built for my thesis, Tanglaw is a multi-pipeline facial recognition system built for low-light environments, utilizing RetinaFace & MTCNN for facial detection, and MobileFaceNet & ArcFace for recognition.

Next.jsFlaskRetinaFaceMTCNNMobileFaceNetArcFace
WebsiteDemo
Bant.ai landing page

Bant.ai

2025

Award-winning digital safety application from Can You HackIT 2025 that uses screenshots, OCR, and LLM-driven moderation to detect and flag harmful content in near real time.

Next.jsElectronFirebaseFlaskOpenAI APIOCR
WebsiteDemo
AlzAware application screenshot

AlzAware

2024

Machine learning application for early Alzheimer's detection using CNN models. Built with PyTorch for accurate image analysis and Next.js for a user-friendly interface, helping healthcare professionals with diagnostic support.

Next.jsFirebasePython FlaskCNN Model (PyTorch)
Website
Aphrodite web application screenshot

Aphrodite

2024

Modern web application showcasing elegant UI/UX design principles. Features responsive design, smooth animations, and efficient data management with MongoDB backend integration.

Next.jsMongoDB
Website
Elysian web application screenshot

Elysian

2023

Full-stack web application built with React and Express.js, featuring comprehensive user management, real-time data processing, and seamless MongoDB integration for optimal performance.

ReactMongoDBExpress.js
Website

Want to get in touch?

Open to full-stack and AI/automation work, collaborations, or a conversation about neuroscience and games. My inbox is the fastest way to reach me.