LE NGOC HIEU

AI Engineer

adw6zo@r.postjobfree.com 082******* Ho Chi Minh City, Viet Nam

OBJECTIVE

To pursue a career as an AI Engineer, where my passion for artificial intelligence can be fully utilized. I am eager to apply my experience in Machine Learning, Deep Learning, Computer Vision, Natural Language Processing to develop innovative solutions that push the boundaries of AI systems and contribute to the growth of the business. I am driven to make a positive impact in the field and am committed to continuously learning and staying current with the latest advancements in AI.

EDUCATION

Ho Chi Minh City University of Technology

Bachelor’s degree: Computer Science

08/2018 – 08/2022

Ho Chi Minh, Viet Nam

Graduated - GPA: 8.03/10

Background AI knowledge: Linear Regression, Gradient Decent, Logistic Regression, KNN, SVM, RNN, LSTM, CNN, Attention, BERT, NLP.

Other languages/tools: Python, JavaScript, C, C++, R, SQL, OS, Statistics, UML. WORK EXPERIENCE

AI Engineer

OLLI AI Technology

Project:

05/2022 – present

Ho Chi Minh, Viet Nam

Maika virtual assistant:

•Description: A Virtual Assistant named Maika for Vietnamese people that allows users to manage it with their voice for entertainment, reminders, music, information search, smart home ownership, and various other functions.

•Responsibilities:

•Research the strengths and weaknesses of several state-of-the-art published pre-trained language models for Vietnamese such as PhoBert, BARTpho.

•Create pre-trained language models with additional spoken language data using DeBerta architecture exclusively for Maika.

•Apply JointIDSF model for intent detection and slot-filling tasks to respond to commands/requires of users.

•Data collection, data generation, training testing, metric evaluation, and data analysis.

•Developing API Backend of Maika system.

•Deploying docker for services on server.

•Techniques: Pytorch/PytorchLightning, Transformers, Pandas, TensorBoard, ONNX, gRPC, Docker.

Internship Odoo Staff

Novobi Company

•Responsibilities: ResDesign and develop web applications on the Odoo ERP platform. 06/2021 – 08/2021

Ho Chi Minh, Viet Nam

•Techniques: Odoo, Python, PostgreSQL

SKILLS

Programing Skills

- Language: Python, JavaScript, C++.

- AI Frameworks/Application: Pytorch, PytorchLightning, NumPy, Scikit-learn, Pandas, OpenCV, SBERT, TensorFlow.

- Visualizations: Matplotlib, Seaborn, TensorBoard.

- Deployment: Docker, ONNX, gRPC.

- Meta-Learning: MAML, ANIL, learn2learn.

- Database: MongoDB, FireBase, PostgreSQL, SQLServer.

- Others: Heroku, Netlify, ReactJS, Django, NodeJS, OpenAI GPT, Prompt, Selenium.

- Language: English

Soft Skills

- Problem-solving

- Creativity

- Communication skills

- Decision marking

- Teamwork

- Curiosity

PROJECTS

Crop Forecasting Utilizing Satellite Data with Temporal-Spatial Transformer Model This AI project leverages spectral bands of satellite data for image reconstruction. Through data preprocessing and a temporal-spatial transformer model, the objective is to enable the AI model to learn the biological characteristics and crop phenology of the field across different seasons. This acquired knowledge is utilized to make accurate predictions of field yields. 03/2023 – 04/2023

Techniques: Transformers, Computer Vision, Pytorch, Pandas Meta-learning for Few-shot Classification on QuickDraw Dataset Applied a meta-learning algorithm called ANIL for a few-shot classification task on the QuickDraw dataset. I conducted experiments with various settings of a few-shot classification model to evaluate the performance of the ANIL algorithm. 12/2022 – 01/2023

Techniques: Pytorch, learn2learn, Meta Learning(MAML, ANIL), CNN. A Jointly Language-Image Model for Multilingual Visual Question Answering Developed a model for the Multilingual Visual Question Answering (mVQA) task at VLSP 2022 – EVJVQA Challenge, which combined T5 and ViT models to achieve superior performance. The key to this success was combining the output feature extractor layer of both models. 10/2022 – 12/2022

Techniques: Pytorch, Transformers, Pandas, Huggingface, ViT, T5. Graduation Thesis: Plagiarism Detection System

Developing a plagiarism detection system specifically for Vietnamese students, with the goal of raising awareness about plagiarism and providing a user-friendly tool to improve writing skills and prevent plagiarism.

06/2021 – 07/2022

Techniques: n-gram, TFIDF, SBERT, FAISS, ReactJS, Django, PostgreSQL, Scikit-learn, Pandas, NumPy.

Team size: 3 - Role: Backend Developer, AI researcher Grade of thesis: 9.93

PUBLICATIONS

Vietnamese Sentence Paraphrase Identification Using Sentence-BERT and PhoBERT Ngoc Hieu Le, Quoc Long Phan, Tran Huu Phuoc Doan,, Ngoc Bao Duy Tran & Tuong Nguyen Huynh DOI: 10.1007/978-3-031-15063-0_40

23/04/2022

OhYeah at VLSP2022 Shared Task: A Jointly Language-Image Model for Multilingual Visual Question Answering.

Ngoc Hieu Le, Luan Dinh Ngo and Quoc Long Phan

06/12/2022

AWARDS

The title "Sinh vien 5 tot" Ho Chi Minh City 2022

Ho Chi Minh City Youth Union

08/01/2023

Top 2 VLSP2022-Multilingual Visual Question Answering VLSP - Association for Vietnamese Language and Speech Processing 06/12/2022

The title "Sinh vien 5 tot" Vietnam National University Ho Chi Minh City 2021 Vietnam National University Ho Chi Minh

10/08/2021

CERTIFICATES

Coursera: Natural Language Processing with Attention Models Coursera: Machine Learning

Certificate of attendance: AWS Innovate - Data and AI/ML Edition REFERENCES

Suong N. Hoang, NLU Team Leader, OLLI

adw6zo@r.postjobfree.com

Contact this candidate