Hi, I’m Cuong>

Key contributions

• Led a team of 4 to build an AI scam website detector with 300+ daily active users, earning 2 national awards: Nhan Tai Dat Viet and Make in Vietnam.
• Fine-tuned custom AI models on ~700K datapoints to screen millions of suspicious URLs, reaching 99.06% accuracy and detecting 200K+ malicious sites in 2 months.
• Built an agentic RAG scam adviser using vector search, keyword search, and web search for 3,000+ users.
• Built platforms for government agencies, including Vietnam Cyber Security Magazine's CyberZ and Digital Trust Alliance clubs in Gia Lai and Hai Phong.
• Developed a "real or fake" awareness game for 25,000+ players.
• Boosted Lighthouse performance from 60 to 100 and scaled legacy APIs to millions of daily requests, secured with reCAPTCHA, Cloudflare, rate limiting, and automated alerts.
• Sole technical point of contact for 10+ international partners.

Key contributions

• Built an enterprise threat intelligence platform for major Vietnamese banks and government agencies, processing terabytes of security signals across 20+ microservices.
• Architected end-to-end data pipelines with automatic crawling and AI analysis to detect impersonation websites, social media mentions, leaked credentials, source code, and dark web incidents.
• Developed an AI agent for analysts to investigate across the platform's databases, with OSINT tools, data exports, and executive DOCX/PDF report generation.
• Implemented tenant isolation, role-based access control, and audit logging.

Key contributions

• Built a science-backed AI vocabulary system for 1,000+ learners.
• Integrated LLMs for personalized test analysis and question answering.
• Implemented a complete admin dashboard for analytics, user, and content management.
• Resolved a critical performance issue for 5x faster page loads.