Henry Vu - Computer Science Graduate Student at UTDallas in Dallas, Texas

Hey, it's Henry

About Me

I graduated summa cum laude from the University of Alberta and I'm currently a CS grad student at UT Dallas. I'm also a founding ML engineer at eXRealityAI, where I'm building agentic voice systems and multimodal XR applications.

I got into math early through competitions and olympiad prep. The contest tricks were fun, but what really stuck was learning how to take a messy problem and find the right angle to crack it. That eventually led me to research at Amii, where I worked on online algorithms and multi-armed bandits, making algorithms robust to worst-case inputs while still taking advantage of good predictions, and studying the exploration-exploitation tradeoffs that come up in bandit problems.

Over time I got more drawn to the gap between theory and practice, building things that actually work on real hardware, which is what brought me to where I am now.

I write occasionally at henryvu.blog.

My Resume

Henry Vu - Dallas Computer Science Graduate Student at UTDallas

Where I've Worked

2025

AR/ML Developer @eXRealityAI

Aug 2025

  • Built voice-to-voice RAG system (Whisper STT → hybrid retrieval with BM25 + FAISS → Mistral-7B-q4 → Kokoro TTS) achieving sub-10s latency with local inference on NVIDIA Jetson Orin AGX.
  • Engineered NLP pipeline with EmbeddingGemma-300m, BAAI cross-encoder reranking, query rewriting, and intent-aware retrieval reducing processing by 40-70%.
  • Integrated Gemini 2.5 with structured JSON schema and context-aware prompt engineering, orchestrating stateful multimodal workflows combining YOLOv9 and Wit.ai voice input across 2 XR applications.
  • Mentored development teams on LLM API integration and prompt engineering for Meta Quest applications.

Computer Vision Engineer @ThorMed Innovation

Feb 2025 - Present

  • Led domain-aligned transfer learning for NIH-funded bladder segmentation project. Pretrained U-Net and SimSiam SSL encoders on 9.2K thyroid/breast ultrasound images, achieving 95.99% Dice on downstream task.
  • Enabled edge deployment via 4-bit PTQ with 7x robustness improvement over ImageNet initialization.
  • Built automated data pipeline with PyTorch and OpenCV: extracted, preprocessed, and segmented 486 clinical images from ultrasound videos with augmentation and quality validation frameworks.

Teaching Assistant @UTDallas

Jan 2025 - Present

  • Held weekly office hours, prepared seminar and exam materials, and graded coursework.
  • Mentored 100+ students in Algorithms and Data Structures through technical reviews.
2024

Undergraduate Research Assistant @SODALab

Dec 2023 - May 2024

  • Reinforcement Learning: Conducted an in-depth theoretical and empirical survey about multi-armed bandit problems under settings such as stochastic, adversarial, Markovian, and restless bandits.
  • Designed and implemented Python simulations to empirically validate and compare the performance of classic bandit algorithms, including UCB, Exp3, Thompson Sampling, and Gittins Index.
  • Supervised by Dr. Xiaoqi Tan.
2023

Teaching Assistant @UAlberta

Sep 2023 - May 2024

  • Held weekly office hours, prepared seminar and exam materials, and graded coursework.
  • Mentored 300+ students in Algorithms and Data Structures through technical reviews and weekly problem-solving sessions.
2022

Undergraduate Researcher - Online Learning @Amii

Apr 2022 - May 2023

  • Online Learning: Researched and implemented algorithms for online optimization problems such as online conversion, knapsack, and bipartite matching using the online primal-dual framework.
  • Studied a new online learning framework that incorporates machine-learned predictions to develop both robust and consistent algorithms, extending beyond traditional worst-case analysis.

Projects I've Worked On

2026

Agentic Framework for Suspect Detection

Multi-agent pipeline using LangGraph and Gemini for automated suspect identification from surveillance data.

Python LangGraph Gemini FastAPI FAISS
2026

BTS Concert Ticket Buying Simulator

Interactive web simulator recreating the BTS concert ticket buying experience with real-time queue mechanics.

HTML/CSS JavaScript Node.js
2026

Valentine Surprise

A fun interactive Valentine's Day web experience with animations and surprises.

HTML/CSS JavaScript
2025

Ultrasound Bladder Segmentation

NIH-funded project: pretrained U-Net encoders on 9.2K ultrasound images, achieving 7x robustness improvement in 4-bit quantization for edge deployment.

SSL OpenCV Model Quantization
2025

Modeling Political Sarcasm

Detection models using feature engineering and fine-tuned DistilRoBERTa to identify subtle political sarcasm in online text.

PyTorch NLTK spaCy scikit-learn
2025

TFT Rolling Odds Calculator

Probability calculator using Markov Chains to compute champion finding odds based on level, pool size, and gold.

HTML/CSS JavaScript Chart.js
2025

Polyps & Breast Ultrasound Segmentation

Attention U-Net for medical image segmentation on polyps and breast ultrasound datasets.

PyTorch OpenCV scikit-learn
2025

A Survey of Geometric Set Cover

Survey of approximation algorithms and PTAS approaches for the geometric set cover problem.

Greedy Approximation Algorithm PTAS
2024

This Web Portfolio

Personal portfolio with theme switching, smooth animations, and responsive design.

HTML/CSS JavaScript
2024

UManitoba Navigator

Hackathon project: campus navigation app with Python FastAPI backend and React frontend.

Python FastAPI React HTML/CSS
2023

ViT and Contrastive Representation Learning

Vision Transformer with contrastive representation learning for image classification.

PyTorch Metric Learning NumPy
2023

VAE and Diffusion for FashionMNIST

Variational autoencoders and diffusion models for generating FashionMNIST images.

PyTorch NumPy
2023

Yahoo's News Recommendation with MABs

Multi-armed bandit algorithms applied to Yahoo's news article recommendation dataset.

Bandits Algorithms NumPy Matplotlib
2023

EEG Decoding: a Multi-Modal Approach

Novel EEG feature extraction combining convolution and self-attention, fused with LLM embeddings for brain signal classification.

PyTorch MATLAB HuggingFace
2022

Gomoku Solver

AI game engine using alpha-beta pruning and minimax search for Gomoku.

Python Alpha-Beta Pruning Minimax
2022

Online Algorithms: Seminar Slides

Research seminar slides on online algorithms and online convex optimization.

Online Algorithms Online Convex Optimization
2021

HabiTrak

Android habit tracking app with Firebase backend and Google Maps integration.

Java Android Studio Firebase Google Maps API
2021

Toronto Neighbourhoods Data Analysis

Spatial clustering analysis of Toronto neighbourhoods using Foursquare API and web scraping.

NumPy Pandas BeautifulSoup Foursquare API
2020

Sudoku Solver

Arduino-based Sudoku solver using backtracking algorithms on embedded hardware.

C++ Arduino
2020

Edmonton Restaurant Finder

Arduino pathfinding system for optimal routes to Edmonton restaurants on a map display.

C++ Arduino
2019

Encrypted Arduino Communication

RSA-encrypted serial communication between Arduino boards using fast modular arithmetic.

C++ Arduino RSA
2019

Huffman Coding for File Compression

File compression tool implementing Huffman coding for efficient data encoding.

Python Huffman Coding

Get in Touch

I read every message. Drop me a line and I'll get back to you soon!