I'm a passionate developer specializing in cloud applications and web development. Welcome to my portfolio where I showcase my projects and skills.
A full-featured workout platform with user authentication, courses enrollment,
track workout with CRUD, fast workout setup for beginner and advanced, chat bot,
week and month analysis, weekly promotions. By using green as it psychologically represents growth, health,
and success—perfect for tracking fitness progress and motivating users toward their goals. The project is using response design,
viewing in mobiel phone is possible especially the course timeable.
There is an admin account for the admin to manage the courses, coaches, users.
Test account: test123456
password: 123456
Admin account: admin123456
password: 123456
(For security reason, when login with admin account, do not Remember me, else it will fali to login)
A simple penetration(.apk) created by msfvenom, in VMWare. It helps me developed a basic concept about the cyber security, how the hacker hack a phone or pc, and help to protect the data.
View on Github
This project proposal was developed during an NVIDIA workshop hosted by HKSTP. A comprehensive proposal for developing a prototype AI-driven phishing detection and mitigation assistant that focuses on email as the primary attack vector. The system parses incoming emails into structured features and analyzes them to produce a phishing risk score. Based on this assessment, intelligent agents decide whether to flag, quarantine, or allow messages, providing automated protection against phishing attacks while minimizing false positives.
View Proposal (PDF)
The deep learning course focuses on how to train AI, which is a subset of machine learning. In this course, I was able to train the model to identify what animal is from a picture by using Python. It became an important foundation for my future AI training and provided hands-on experience with neural networks and model optimization.
View Certificate
I learned the concept of NVIDIA's Transformer architecture. It can transfer the user's natural language into numeric tokens, and then convert them into vectors (matrices). These matrices are processed through multiple layers of neural networks using attention mechanisms to understand context, relationships, and meaning. Finally, the model generates a response by predicting the next most likely token in sequence. It gave me a huge concept about how AI works and will help my future work related to AI engineering, particularly in building AI models.
View CertificateWorkout App Screenshot