>

I build full-stack AI apps, from agentic RAG systems to scam-detection tools, serving thousands of real users.

AI & Full-stack Web

Tech Lead atChongLuaDao

Ho Chi Minh City, Vietnam

Get in touch See my CV

Experience

Tech LeadatChongLuaDao

Jan 2025Present

Key contributions

Led a team of 4 building an AI scam website detector with 300+ daily active users; won 2 national awards (Nhan tai Dat Viet and Make in Vietnam).
Integrated frontier LLMs with a LoRA fine-tuned open-source model to balance cost and performance.
Architected an agentic RAG system for a scam adviser AI combining vector-based search, keyword search, tool calling, and real-time web search, reaching over 3,000 users.
Shipped 3 full-stack web apps across a VPS network.
Optimized legacy frontend code to boost Lighthouse performance from 60/100 to 100/100.
Scaled legacy APIs from thousands to millions of requests/day, hardened with reCAPTCHA, Cloudflare, and rate limiting, paired with monitoring & automatic alerts.
Served as the sole technical point of contact for 10 international partners.

Tech stack

Next.js
SolidJS
TypeScript
Python
PostgreSQL
Redis
MinIO
Docker

Reference letter

From Ngo Minh Hieu (Hieu PC) – CEO of ChongLuaDao

Products shipped

ai.chongluadao.vn

AI scam website detector that classifies threats with LLMs, explores sites via a live browser agent, aggregates threat intelligence reports, and extracts identifiers (emails, phones, socials, crypto wallets).

zalo.me/chongluadao

Zalo chatbot that checks websites, images, and phone numbers scam signals, and answers safety questions based on ChongLuaDao's knowledge base.

tracnghiem.chongluadao.vn

A game where players spot real vs. fake websites, emails, messages, videos, and images.

baovethongtin.com

Web app that improves users' digital security via personalized checklists, assessments, and progress tracking.

Full-stack Developer & AI IntegratoratLumist

MayAug 2025

Key contributions

Built an AI-powered, science-backed vocabulary learning system for 1,000+ users.
Integrated LLMs for personalized test performance analysis & question answering.
Implemented a complete admin dashboard for data analytics, user & content management.
Resolved a critical performance issue to achieve 5x faster page loads.

Tech stack

Next.js
TypeScript
Supabase
Vercel AI SDK

Product shipped

lumist.ai

An all-in-one Digital SAT prep platform with an extensive question bank, AI-driven performance analysis, predictive scoring, gamification, and more.

Projects

Examinai

AI-powered IELTS Learning Platform

Examinai

AI-powered IELTS Learning Platform

Integrated AI models across numerous modalities, including text generation, image understanding, text-to-speech, and speech-to-text.
Architected writing and speaking assessment flows combining multiple LLMs, with official marking criteria, structured outputs, response streaming, and retry logic.
Developed an agentic RAG system with an IELTS knowledge base for question answering.
Curated more than 1,000 practice tests via web scraping.

Tech stack

Next.js
TypeScript
Python
Supabase
Vercel AI SDK

See it in action GitHub

InfiniteCompute

E-commerce Store for GPUs with AI Advisor

InfiniteCompute

E-commerce Store for GPUs with AI Advisor

Built a fully-featured web with order tracking, reviews, coupons, authentication, admin panel, etc.
Created an agentic RAG system for GPU recommendations combining vector-based search, product database interaction via tool calling, and real-time web search.
Containerized all components via Docker to support easy reproducibility across environments.

Tech stack

Vue.js
Python
FastAPI
PostgreSQL
Docker
OpenRouter

See it in action GitHub

MatchFluence

Find Influencers using AI

MatchFluence

Find Influencers using AI

Developed the entire platform with a full database and AI pipeline in 3 days for a hackathon.
Curated a dataset of 2,000 influencers using web scraping.
Designed a recommendation system combining SQL filtering and AI evaluation.

Tech stack

Next.js
TypeScript
Python
Supabase
Vercel AI SDK

See demo video

Research

Published

A Novel Spam Review Detection Architecture Using Large Language Models

Spam review detection architecture: ViSpamReviewV2 dataset feeding a preprocessing pipeline into the Llama 3.1 model

Fine-tuned 3 LLMs using LoRA to detect spam reviews on Vietnamese e-commerce platforms.
Improved accuracy of prior work from 88.54% to 92.12%, and F1 score from 73.49% to 78.86%.
Analyzed and compared model architectures and sizes in detail (PhoBERT vs Llama 1B, 8B, 70B).
Deployed models using serverless GPU inference for scalability and cost-effectiveness.

Read the paper

Under Review

Beyond URLs: Content-Aware Malicious Website Detection Using Multilingual BERT

Malicious website detection pipeline: four datasets feeding a combined dataset, split into URL and HTML features, evaluated by BERT and traditional ML models

Trained 20 model configurations across transformers, traditional ML models, and hybrid ensembles, reaching 99.06% accuracy on a 68,954-sample test set.
Surpassed results of 4 prior works, improving accuracy by up to 8.28%.
Compiled and cleaned a large-scale dataset of 689,556 webpages from four real-world sources.
Designed an HTML-to-Markdown preprocessing pipeline to reduce input length by over 96% while preserving key signals.
Conducted parallel training on a network of cloud-based GPUs, optimizing cost and performance.

Read the paper

Recognition

Awards

Nhan Tai Dat Viet 2025

3rd place, "Promising Digital Technology" category. Top 17 of 122 products at Vietnam's most prestigious tech award.

Make in Vietnam 2025

Bronze in "Promising Digital Technology". Jumped from Top 10 (2022) to Top 3 with the new AI-powered scam website detector.

Best Developers in Vietnam 2023

3rd place against 400+ participants in a nationwide hackathon with our AI study companion project.

LeetCode T-shirt

A badge of honor among LeetCoders. Earned through solving 750+ problems on data structures and algorithms throughout 2 years, ranking in the top 5%.

Vietnam Television (VTV) coverage

Dec 2025

VTV9 nationwide anti-scam panel

Panelist on VTV9's nationwide anti-scam campaign with Viettel Security and the Ministry of Public Security, sharing the tech perspective alongside policy and law-enforcement experts.

Sep 2025

VTV9 talk show: AI-powered scam prevention

Featured guest on VTV9's Kính Đa Chiều talk show, breaking down how AI can be used to detect and stop online scams targeting Vietnamese users.

Dec 2025

VTV9 talk show: Living in the age of AI

Returning guest on Kính Đa Chiều to discuss the broader impact of the AI era, from everyday productivity to the new risks it brings.

Mar 2026

VTV9 talk show: Inside AI-powered scam tactics

Invited back to Kính Đa Chiều to walk viewers through emerging AI-driven scam techniques like deepfakes, voice cloning, and AI-generated phishing.

Presentations

Bangkok, ThailandAug 2025

Mekong-U.S. Partnership Track 1.5 Policy Dialogue on Digital Economy

Joined the Strengthening Cyber Resilience in ASEAN panel alongside Microsoft and Cloudflare representatives to share how Chongluadao detects AI-generated scam websites at scale.

Hue, VietnamSep 2025

Security Bootcamp 2025 by Vietnam Internet Association

Presented "Applying AI in Cybersecurity: Detecting Scam Websites Using LLMs" with the Chongluadao AI team at Vietnam's flagship infosec event, themed "Resilience."

Can Tho, VietnamOct 2025

FPT International Conference on Emerging Trends in Computing 2025

Presented our LLM-based spam review detector for Vietnamese e-commerce at FETC's inaugural edition, themed "AI Solutions for Developing Countries."

OnlineSep 2025

Young Southeast Asian Leadership Initiative (YSEALI) Roundtable

Invited back by the Stimson Center as an expert speaker to share key takeaways on cyber resilience and scams with YSEALI alumni ahead of the Policy Dialogue's summary report launch.