Team
LLM-Freshman
Project Concept
No description has been added yet.
Entry
Status: Submitted
Last saved: May 09 at 5:18 PM HKT
Team Roster
You must be registered for the event to view the team message board.
MAK YAT LONG Team Lead RSVP Approved
Student @ SBC | Independent Al Cybersecurity Researcher | Al Safety & LLM Red Teaming at St. Bonaventure College & High School
Solo, I built most of the product through Replit.
An independent AI Cybersecurity Researcher and a Form 4 student at SBC. Specializes in adversarial prompt engineering and LLM safety. Research has been recognized by industry leaders at HKGAI and Votee AI. Successfully reported validated vulnerabilities to Google’s AI VRP, including an IPI discovery in Gemini. Passionate about securing the future of AI-driven interfaces.
Interested in learning more about the A2UI and AG-UI protocols and how they handle state and security during dynamic rendering. Eager to connect with senior engineers specializing in Agentic UX and RAG security to explore how to make 'self-healing' secure interfaces.
Researching Indirect Prompt Injection (IPI) and UI Redressing in Generative UI. Currently building a security layer for Agentic Interfaces that uses LLM-based self-auditing to prevent data exfiltration via malicious MCP servers. Also maintaining a vulnerability research portfolio focused on Edge AI safety.