Offensive Security Engineer & AI Security Researcher
Red Teaming • AI Security • Web/AppSec

I am an Offensive Security Engineer and AI Security Researcher with hands-on experience across enterprise penetration testing and applied AI and security research. I deliver end-to-end assessments across Web, API, Mobile, and Network environments, and contribute to Red Team operations focused on real-world attack scenarios, identity abuse, and security posture improvement. My work also includes security automation, critical vulnerability response support, and AI security readiness for AI-enabled applications. I conduct applied identity and access security work on Microsoft 365 Conditional Access policy enforcement, with tooling selected for Black Hat Asia 2026 (Arsenal). Open to global opportunities in Red Teaming, Identity Security, and AI Security.

Key Responsibilities:
Skills: Penetration test · Kali linux · Burp Suite · AI Security · Reporting · Power Automate · Power Apps


Key Responsibilities:
Skills: Penetration test · Kali linux · Burp Suite · OWSAP · Vulnerability Assessment


Key Responsibilities:
Skills: Penetration test · Kali linux · Burp Suite · OWSAP · Cybersecurity · Vulnerability Assessment


Key Responsibilities:
Skills: Android Development · OAuth2.0 · Application Security · Security · Cybersecurity · Java


Key Responsibilities:
Skills: Public Speaking · Jupyter · Convolutional Neural Networks (CNN) · Deep Learning · Artificial Intelligence (AI) · Communication · Python
News Conference Paper Show Credential Conference Certifications

Activities and societies:

Activities and societies:
Activities and societies:



Research papers presented at international IEEE conferences
IEEE Region 10 Conference 2024
Poultry farming is crucial to the food chain, and chicken health directly impacts product quality and safety. Diagnosing poultry diseases using polymerase chain reaction is costly, particularly for small farms. To address this, we developed a mobile-based service for farmers, enabling the identification of common chicken diseases from fecal images via a Line account. Our system achieved 86.49% segmentation precision and 95.93% classification accuracy on a large dataset, offering a practical and accessible tool for local farmers.
23rd IEEE International Conference on Software Quality, Reliability, and Security
OAuth 2.0, commonly used for authorization, can be susceptible to CSRF attacks in Android applications. To address this, we developed an Android app to assess other apps' use of the OAuth 2.0 state parameter—a key defense against CSRF. Our analysis, conducted on both Chrome and the default browser, evaluates whether Android apps using OAuth 2.0 are adequately protected against CSRF attacks. Our research aims to protect users by highlighting apps with potentially vulnerable OAuth 2.0 implementations.
6th International Conference on Information Technology
An automated COVID-19 screening framework using chest X-ray images is proposed in this study. It leverages artificial intelligence techniques and transfer learning for accurate diagnosis. The framework extracts features using transfer learning and applies modified deep neural networks. Grad-CAM visualization supports the predicted diagnosis. Experimental results demonstrate superior performance compared to other deep learning techniques. This framework has the potential to aid in early COVID-19 diagnosis and alleviate the burden on radiologists.
The TableMaster for Elementor plugin for WordPress is vulnerable to Server-Side Request Forgery in all versions up to, and including, 1.3.6. This is due to the plugin not restricting which URLs can be fetched when importing CSV data from a URL in the Data Table widget. This makes it possible for authenticated attackers, with Author-level access and above, to make web requests to arbitrary locations, including localhost and internal network services, and read sensitive files such as wp-config.php via the 'csv_url' parameter.
The LA-Studio Element Kit for Elementor plugin for WordPress is vulnerable to Administrative User Creation in all versions up to, and including, 1.5.6.3. This is due to the 'ajax_register_handle' function not restricting what user roles a user can register with. This makes it possible for unauthenticated attackers to supply the 'lakit_bkrole' parameter during registration and gain administrator access to the site.
The SearchWiz plugin for WordPress is vulnerable to Stored Cross-Site Scripting via post titles in search results in all versions up to, and including, 1.0.0. This is due to the plugin using `esc_attr()` instead of `esc_html()` when outputting post titles in search results. This makes it possible for authenticated attackers, with contributor-level access and above, to inject arbitrary web scripts in post titles that will execute whenever a user performs a search and views the search results page.
Oblivion Token is a focused offensive-research utility that makes Microsoft 365 Conditional Access Policy (CAP) bypass testing practical, repeatable, and scriptable. Designed for red-teamers and security researchers, it systematizes token-centric workflows and exercises policy edge cases to expose where device, network, or app-context assumptions fail including realistic test paths against Microsoft first-party (built-in) applications.
TBA
TBA