Hello, I'm Hao.

I'm a Senior AI Researcher & Engineer specializing in LLM systems, AI engineering, agentic workflows, and real-world machine learning.

With a background in Applied Mathematics and a passion for AI, I share daily insights on LLM systems, AI interview questions, system design, and cutting-edge research papers. My work focuses on the intersection of research and engineering, turning cutting-edge papers into production-ready AI systems.

I write about AI interview questions (LLM engineering, system design, research), LLM infrastructure & optimization, RAG architectures (GraphRAG, Agentic RAG, Hybrid), Vision-Language Models (VLMs), and building AI products at scale.

I can help you out with:

AI Development
AI Research
Python Backend Development
GCP Infrastructure
and more...

Follow me on LinkedIn Subscribe to Substack

🚀 45.8K+ LinkedIn followers

📬 5.5K+ Substack subscribers

🇻🇳 Top #1 AI & Machine Learning in LinkedIn Vietnam

🇻🇳 Top #3 IT & Tech in LinkedIn Vietnam

🇻🇳 Top #10 LinkedIn Vietnam

🌍 Top #20 AI & Machine Learning LinkedIn Worldwide

👋

Check out my projects

My Projects

Here are some of my recent projects. I'm always working on something new, so check back often!

🎙️ AI Travel Guide Voice Bot A sophisticated real-time AI travel guide powered by AssemblyAI, OpenAI, and ElevenLabs. This intelligent voice bot provides personalized travel recommendations and answers questions with natural-sounding AI voice.

🧠 DeepSeek Thinking UI A professional ChatGPT-like interface with visible reasoning powered by DeepSeek-R1. Experience AI thinking process in real-time with an intuitive user interface.

⚡ LLM Quantization Zoo A collection of hands-on notebooks demonstrating various model quantization techniques, including GPTQ, GGUF, ExLlamaV2, and 8-bit quantization for efficient model deployment.

📚 LLM Finetuning Cookbook A hands-on, code-first cookbook for fine-tuning Large Language Models (LLMs) across various techniques and tasks. No frills, just practical code and real-world applications.

🎯 Ensemble Learning Tutorial A comprehensive machine learning project demonstrating ensemble learning techniques including bagging, boosting, and stacking methods with Jupyter notebooks and Docker support.

👁️ Attention Captioning Tutorial PyTorch implementation of image captioning with and without attention mechanisms using Flickr8K dataset. Learn the fundamentals of computer vision and natural language processing.

🔧 Deep Learning Optimization Tutorial Interactive deep learning optimization tutorial with Jupyter notebooks. Master neural network tuning, hyperparameter optimization, and model performance techniques through practical exercises.

📈 Trading Floor Agent Production-ready autonomous trading floor with 4 AI agents making trades, powered by 6 MCP servers and 44 tools. Features real-time dashboard, portfolio analytics, and Docker support.

🤖 Agent Creator AI agent that creates other AI agents using AutoGen. Features dynamic agent generation, gRPC runtime, Streamlit UI, and Docker support for scalable agent communication.

🦸 Sidekick Agent LangGraph-powered personal AI co-worker that works in your browser. Can browse the web, run Python code, search Wikipedia, manage files, and more - your own version of OpenAI's Operator Agent.

👷 Engineering Team Agent 4-agent CrewAI system that designs, builds, tests, and creates UI for software modules. Takes high-level requirements and produces complete, tested solutions with Docker support.

📊 Stock Picker Agent AI-powered investment research and stock selection agent built with CrewAI. Uses specialized agents to find trending companies, conduct financial research, and select investment opportunities.

🔍 Deep Research Multi-agent research system that conducts comprehensive research on any topic. Orchestrates specialized agents to plan searches, gather information, synthesize reports, and send results via email.

📧 SDR Agent AI-powered Sales Development Representative that crafts and sends professional cold emails. Uses OpenAI to generate personalized emails and supports bulk sending via SendGrid or SMTP.

👤 Career Digital Twin AI agent that represents you to potential employers. Uses your resume, LinkedIn profile, and personal summary to answer questions about your career, skills, and experience.

View All My Projects

Some of my writing

My Writings

Along with coding I also like to write about life and technology. Here are some of my recent posts.

Advanced Reinforcement Learning Interview Questions #6 - The Initialization Gap Trap

You’re in a final-round interview for a Senior AI Engineer role at NVIDIA Robotics .

Posted on February 1, 2026

Advanced Reinforcement Learning Interview Questions #5 - The Success-Only Dataset Trap

You’re in a Research Scientist interview at Google DeepMind , and the lead researcher throws you a curveball: “I have a dataset of reasoning traces, but they’re all flawed.

Posted on January 31, 2026

Advanced Reinforcement Learning Interview Questions #4 - The LLM-as-a-Judge Trap

You’re in a Machine Learning interview at DeepSeek AI and the lead researcher asks: “We want to train a reasoning model using 𝐃𝐢𝐫𝐞𝐜𝐭 𝐏𝐫𝐞𝐟𝐞𝐫&#1...

Posted on January 30, 2026

View All My Writing

Subscribe to my newsletter

Get daily AI interview questions, LLM system design insights, and research paper breakdowns delivered to your inbox.

Subscribe on Substack

© 2026 Aria