HUNG LE-QUANG

ad30du@r.postjobfree.com https://github.com/qhle2001 098******* HCMC, VN

OBJECTIVE

* ****** ** ********** ** NLP, the focus has been on Large Language Models. Use LangChain, and Openai to create products. In addition, trying to apply some other models to the products. EDUCATION

University of Information Technology – Ho Chi Minh City National University September 2020 - Present Major: Computer Science

GPA: 8.11

Expected graduation date: June 2024

TECHNICAL SKILLS

Python: Working with FrameWorks Machine Learning, Artificial Intelligence. For example: Tensoflow, Keras, Numpy, Scikit-learn, OpenCV, Convolution Neural Network.

Other skills: In addition, I can work with C#, Java, HTML/CSS, Javascript, C/C++, and MySQL. LANGUAGE

English: Effective communication

EXPERIENCE

AI Engineer Internship – Hephaestus Technology October 2023 – February 2024

Support creating chatbot with some files such as csv file. Use LangChain to generate a description of file uploaded for client, fix chart, and try some models (such as Mistral7B, Codellamma) to apply in this chatbot.

Use Openai and ImageMagick to create VideoGeneration tool for internal use. Click here to watch its output.

Research some chatbot can visualize data.

PROJECTS

Dropout Student Prediction July 2023 Python

Description: Construction of a model to predict whether a student’s dropout or not used Linear Regression, Decision Tree, and Support Vector Machine.

Technical used: Linear Regression, Decision Tree, Support Vector Machine, Numpy, Pandas, Sklearn, Matplotlib, XGBost, …

Main tasks:

- Explore Data Analysis.

- Removing features from dataset that were not necessary.

- Training model with algorithms above (Linear Regression, Decision Tree, Support Vector Machine).

- Training with Hyperparameters tuning (GridSrearch).

- Comparing those trained models’ results.

Vietnamese Sentiment Analysis July 2023 Python

Description: Construction of a model to predict attitude of customer commets about the restaurant and building a corpus.

Technique used: Linear Regression, Support Vector Machine, Pandas, Numpy, Sklearn, Unicodedata, Emoji, …

Main tasks:

- Preprocessing dataset.

- Training model with Linear Regression and Support Vector Machine.

- Reading and researching the guidelines.

- Annotation of the dataset after prediction from the tranined model (Linear Regression model).

- Comparing the annotators’ results with goal dataset. Human Facial Emotions Recognition February 2023 Python

Description: Building a model to predict human facial emotions using Deep Leaning and Machine Learning techniques.

Technique used: Convolution Neural Network, VGG16, VGG19, Support Vector Machine, Tensorflow, OpenCV, Matplotlib, Sklearn, …

Main tasks:

- Searching and researching those models will be using.

- Finding and collecting the dataset from kaggle.

- Training the models.

- Comparing those models’ results.

Retrieval Information in English Sentence December 2022 Python

Description: Vector Space and Binary Independent model were 2 main methods in project and used to Cranfield data.

Technique used: Pandas, Numpy, Sys, NLTK, Operator.

Main task:

- Preprocessing: remove specific charaters, split words, remove stopwords, stemming.

- Building Binary Independent and Vector Space model.

- Comparing the results of 2 models and compared with whoosh’s results. Facemask Detection June 2022 Python

Description: Used YOLOv4 and YOLOv5 to train the dataset collected from kaggle with detection human facemask purpose.

Technique used: YOLOv4-tiny, YOLOv5, Pytorch, OpenCV.

Main tasks:

- Searching and collecting the dataset from kaggle.

- Searching and studying YOLO family.

- Selecting the YOLO version to train the model.

- Comparing these models’ results and finding a way to improve these models’ accuracy (enhanced dataset). Labeling Vietnamese Words December 2021 to January 2022 Python

Description: Used Longest Matching method to split Vietnamese sentences into words and then used Hidden Markov model to train model.

Technique used: Pandas, Sklearn, NLTK, Pyvi, Underthesea.

Main tasks:

- Created a dataset.

- Split Vietnamese sentences into words.

- Training models use Hidden Markov Model.

- Comparing this model’s results with some libraries.

Contact this candidate