❧
About
Hey, I'm Henry. I'm a machine learning engineer at eXRealityAI in Dallas, and I'm finishing my M.Sc. in Computer Science at UT Dallas, expected Summer 2027. Before that I did my B.Sc. Honors in Computing Science at the University of Alberta.
At eXReality I build XR Auto Service and Smart DVI, a spatial AI platform for auto shops. It puts an agent on a technician's headset that guides them through a repair or an inspection and answers questions from a repair knowledge graph. I've also spent a lot of time getting voice and vision models to run on a Jetson and on Apple silicon, and I shipped GardenXR to the Meta Horizon Store.
On the research side I work with Dr. Katherine Brown at UT Dallas on NIH-funded ultrasound segmentation, and earlier I did online algorithms and bandits at Amii. I write about all of this at henryvu.blog.
My resume
❧
Where I've worked
-
Founding Machine Learning EngineereXRealityAI
Aug 2025 to now
- Built XR Auto Service end to end, a Meta Quest 3 app that guides a technician through a repair, from the headset app and the agent behind it to the knowledge graph, retrieval, and the evaluation harness in CI. Piloted with an OEM; on an expert-written test set it puts the correct procedure first 92 percent of the time.
- Took Smart DVI to production with auto-repair shops: a Meta Quest 3 inspection app that grades every photo with AI, runs agents on the headset, takes voice notes offline, scans VINs, reads OBD-II over Bluetooth, and signs in with a hardware-backed device key. Then read its production data to find where technicians' findings were getting lost, and shipped the fixes.
- Built a voice-to-voice assistant that runs entirely on an NVIDIA Jetson Orin: Whisper for speech in, hybrid BM25 and FAISS retrieval, a quantized Mistral 7B, and Kokoro for speech out.
- Multimodal XR workflows on Gemini, YOLOv9, and Wit.ai voice input across two applications, and a retrieval pipeline (EmbeddingGemma, cross-encoder reranking, query rewriting) that cut retrieval work by roughly half.
- Shipped GardenXR to the Meta Horizon Store, traced a run of bad production answers to a vendor proxy that was silently dropping images, and mentored the team on LLM integration.
-
Computer Vision EngineerThorMed Innovation
Feb 2025 to now
- Modeling lead on an NIH-funded project to segment the bladder in clinical ultrasound, so patients can be monitored without a catheter.
- Pretrained U-Net and SimSiam encoders on 9.2K ultrasound images before fine-tuning, reaching 96 percent Dice.
- Got the model onto portable hardware with 4-bit quantization, where the domain-pretrained weights held up about seven times better than ImageNet weights, and built the pipeline that extracts and cleans frames from clinical ultrasound video.
-
Jan 2025 to May 2025
- Algorithms and data structures: weekly office hours, seminar and exam material, and technical reviews for more than a hundred students.
-
Undergraduate Research AssistantSODALab
Dec 2023 to May 2024
- A theoretical and empirical survey of multi-armed bandits across stochastic, adversarial, Markovian, and restless settings, with Dr. Xiaoqi Tan.
- Simulations comparing UCB, Exp3, Thompson sampling, and the Gittins index.
-
Sep 2023 to May 2024
- Algorithms and data structures for a few hundred students: office hours, weekly problem-solving sessions, and grading.
-
Undergraduate Researcher, Online LearningAmii
Apr 2022 to May 2023
- Online optimization with the primal-dual framework: conversion, knapsack, and bipartite matching problems where decisions are made before all the input arrives.
- Algorithms that use machine-learned predictions to beat worst-case bounds when the predictions are good, without losing the guarantees when they are not.
❧
❧