Ramon PacilonaRamon Pacilona

As Lead Programmer for ADDOS, a cascaded ML based DDoS mitigation system for Software Defined Networks, I trained the hybrid detection models and built the backend pipeline and mitigation logic. I trained the Stage 1 Isolation Forest anomaly detector and Stage 2 Random Forest classifier for SYN, UDP and ICMP floods, and developed the 5 stage Python pipeline with flood prefiltering, entropy analysis, cascaded inference and state machine response integrated with the Ryu controller.

Studium is a web app that turns study notes into a complete study guide. Users upload a PDF, Word, or text file, and the app creates a summary, key terms, topic breakdowns, a concept map, and practice quizzes with explanations. It can generate up to 100 mock quizzes per reviewer, so students can practice until they feel confident. Built with Next.js, React, and Tailwind CSS.

Built my personal portfolio website using Laravel, Inertia.js, and React. Features a custom admin dashboard with complete CRUD functionality to manage projects, experiences, skills, and contact messages in real time.

Served as the backend developer for an academic group e-commerce project. Built the core application logic using Laravel and MySQL, focusing on secure OTP user authentication, database management, and full CRUD endpoints for products, user profiles, and administrative controls.

Pity Tracker, a lightweight and fully client side probability calculator for gacha games. It helps players estimate how many pulls are needed to reach 50 percent, 80 percent, and 95 percent confidence across real banners by modeling base rates, soft pity ramping, hard pity guarantees, 50/50 mechanics, and existing pity offset