N G U Y E N X U A N T U Y E N
AI ENGINEER (JUNIOR)
CONTACT INFO
E-mail adz7qv@r.postjobfree.com
Phone Nr 039*******
Address */** **** **** ******, 2 Ward, Tan Binh District, Ho Chi Minh City
LinkedIn www:linkedin:com/in/tuyen-nguyen-67474921b ABOUT ME
I love to research science, learn new things and it will be even better if I can apply what I have learned to life and work. Now I am researching about AI, and in next 2 years I want to become a senior AI engineer. Therefore, I hope to develope my skills and grow to become a good AI/ML engineer in the near future
My hobbies • Soccer, Music, Game, Math, Universe, Physics EDUCATION
Ho Chi Minh City University of Technology 2018-2022 Major: Control Engineering and Automation Ho Chi Minh city, Vietnam
• GPA: 7.5
Certificates 2022
Coursera
• Introduction to Containers w/Docker, Kuburnetes, Openshift
• Machine Learning Specialization (3 courses)
• DeepLearning.AI TensorFlow Developer (4 courses) ACHIEVEMENTS
Awards • First prize in the 7th Delta International Smart and Green Manufacturing Contest
• Co-author of a scientific paper published on IEEE: Autonomous Mobile Robot: Navigating and Moni-
toring Fire Safety at Power Substation ”https://ieeex- plore.ieee.org/document/9984088”
QUALITIES
Having possitive attitude and initiative to work
Having ability to work under pressure
Having responsibility for teamwork
Strong in researching new knowledge if necessary and working depen- dently
SKILLS
Expertise Skills 2022
• Python, C++
• Libraries: Numpy, OpenCV, Detectron2, Mediapipe, Dlib, Matplotlib, Pandas. Pillow, Scipy, Scikit-Image, ...
• Frameworks: Pytorch, Tensorflow, MXNet, experience in using ONNX
• Deep Learning models: CNN, VGG, ResNet, SSD, YOLO, Fast-RCNN, ...
• Basic knowledge of Machine Learning: PCA, Linear Regression, Logistic Regression, K-mean Clustering, K-mean Neighbors, SVM, ...
• Image processing skills: Edge detection (Canny), Straight lines detection
(Houghlines), Erosion, Dilation, ...
• Some skills in design patterns
• Git, Azure DevOps
• Docker, Kubernetes
• MongoDB
• FastAPI
• Linux bash/shell
Languages • English (780 TOEIC)
EXPERIENCE
Almost 2 years working experience 2022
OCR January 2022
- July 2022
Truong Minh Thinh technology joint stock company
• Receipt information extraction (Use PaddleOCR for detecting text and key information extraction, VietOCR for recognizing Vietnamese text), write pipeline with main modules (pre-process and post-process)
• Book information extraction: Use document layout method for detecting titles, removing unnecessary figures, tables and use Tesseract for recog- nizing text in scanned image (convert from PDF)
FACE RECOGNITION SYSTEM August 2022
- Present
Truong Minh Thinh technology joint stock company
• Training face embedding model, coding face normalization and align- ment module, using Qdrant search engine to match to suitable face
• Faces data augmentation: Use 3D face reconstruction to create face mask then rotate this mask in 3 dimesions (pitch, roll, yaw), extract face texture of real face and combine with created mask by PRNet IMAGE GENERATING September
2022 - Present
Truong Minh Thinh technology joint stock company
• Trial MidJourney for image generation
• Diffusion models for image generation Training Stabe Diffusion (use Dreambooth method) for virtual try on clothes), understand and apply controlnet, inpainting, outpainting, pix2pix for text2img or img2img tasks
• Knowledge and experience with VAE, GAN
IMAGE SEARCH March 2023
- Present
Truong Minh Thinh technology joint stock company
• Image-to-image and text-to-image search: (Use OpenAI-CLIP) Training on a variety of (image, text) pairs. Mapping images and text into the same latent space, use text or image to find similar images HUMAN LANDMARKS DETECTION 2022
Truong Minh Thinh technology joint stock company
• Pose Estimation for Trying on clothes (use Mediapipe, Detectron2
• Face Landmarks Detection for Face Alignment task of Face Recognition system (Use Mediapipe, Dlib
OBJECT DETECTION - INSTANCE SEGMENTATION 2021 - 2022 Self-study projects at school
• Label and train Yolov5 for detecting shrimp (this is a task of contest prod- uct of 7th Delta international Smart and Green Manufacturing Contest)
• Train Unet for segmenting defect on product surface (Self-study project)
• Use Yolov5 for shop signboards segmentation then extracting information from these signboards (OCR)
SKILLS GAINED AFTER WORKING 2021 - Present
• Build and train deep neural networks, identify key architecture parameters, implement vectorized neural networks and deep learning to applications
• Train test sets, analyze variance for DL applications, use standard tech- niques and optimization algorithms, and build neural networks in different Frameworks
• Understand how to deploy and use an AI application and optimize re- sources
• Experience working with GPUs, hardware devices, camera