AI · Computer Vision · Intelligent Systems
Applied AI · Video Understanding · Edge AI · Intelligent Systems
I build and study computer vision systems for real-world applications, with a focus on video understanding, efficient deep learning, object detection, person re-identification, and deployable AI pipelines.
Profile
AI undergraduate with hands-on experience in computer vision, video understanding, object detection, person re-identification, and deployable deep learning systems.
Experienced across research and production-oriented workflows, including dataset preparation, model training and evaluation, real-time video pipelines, ONNX/Docker deployment, and technical reporting. Contributed to peer-reviewed research and AI/robotics projects, with a strong interest in building reliable perception and intelligent systems for real-world applications.
01/2026 – 04/2026
FPT Software - AI Valley
Can Tho, Vietnam
Contributed to production-oriented AI workflows for enterprise use cases, including preprocessing, model training and optimization, inference setup, and validation. Participated in deployment and integration workflows using ONNX, Docker, Linux environments, and REST APIs.
10/2024 – 08/2025
FPTU AI & Robotics Challenge 2025 / Vietnam Open Robotics Challenge 2024
Can Tho, Vietnam
Supported teams with AI problem solving, ESP32, sensors, robotics hardware, connectivity, and on-site troubleshooting during competition deployment.
Technical foundation
PythonC/C++ fundamentals
NumPy
PandasData preprocessingStatisticsModel evaluation
PyTorch
TensorFlow / KerasCNNsAttention mechanismsLSTMTemporal modeling
OpenCVYOLO-based detectionImage preprocessingVideo preprocessingObject trackingPose estimationPerson re-identificationVideo understanding
DockerREST API integration
Linux
GitJupyterGoogle ColabPeer-reviewed work
Concurrency and Computation: Practice and Experience · Wiley
DOI: 10.1002/cpe.70811
Developed a spatial-temporal video recognition pipeline combining YOLOv11-based person detection, AFNet feature learning, and TemporalSA-MoViNet for efficient violence recognition.
2026 IEEE 3rd International Conference on Big Data Science and Engineering (ICBDSE)
Developed a lightweight YOLOv11-based detector using GhostConv and C3k2Ghost modules, targeting efficient inference for resource-constrained vision systems.
HUMAN 2026
Worked on efficient aerial tiny person detection by integrating high-resolution and lossless feature representation for small-object detection.
FPT University · AI / Computer Vision Researcher
Built an end-to-end video analysis pipeline using person detection, feature extraction, attention, and temporal modeling for CCTV footage. Contributed to dataset construction and cleaning, model implementation, experiment design, evaluation, reporting, and manuscript revision.
FPT University · Research Member
Enhanced OSNet with MDHA and Dynamic Graph Convolution; evaluated performance on Market-1501 and DukeMTMC and conducted ablation studies. Contributed to module design, experiments, and technical writing. Related multi-camera tracking work was presented at Research Festival 2025.
2023 – 2027 (expected)
FPT University - Can Tho Campus
Can Tho, Vietnam · GPA 3.5 / 4.0
01/2024
Kasem Bundit University
Thailand
Recognition & learning
Consolation Prize
FPT University
Online multi-camera people tracking with MDHA Attention and DGC Convolution
Specialization
DeepLearning.AI / Coursera
Certificate
Coursera / IBM
Certificate
Coursera
2nd Prize
FPT University
Community
President / Former Academic Chair
FPT University - Can Tho Campus
Managed club operations and technical activities; designed AI/ML training content, guided mini-projects, and coordinated AI/robotics workshops and competition support.
Program Manager / Human Resources Lead
Vietnam
Coordinated youth programs related to sustainability, green jobs, and capacity building, including cross-functional planning across people, communications, and event operations. Contributed to initiatives that delivered 30+ capacity-building workshops and directly benefited approximately 1,400 youth.
Vice President
Can Tho, Vietnam
Let's connect
I'm open to AI internships, computer vision roles, research opportunities, and collaborations on deployable intelligent systems.
Email me© 2026 Thanh Duy Huynh
Customized from the developer-portfolio template.