AI/ML Engineer

PHAN ANH

HCMC • **********@*****.*** • (**+) *** 964 333 • linkedin.com/in/nhật-anh-phan

github.com/PhanAnh-DS-AI

I'm a final-year Analytical Data Science student skilled in Computer Vision, Statistics, Python, R, C#, and SQL. Proficient in Machine Learning and Deep Learning, I'm eager to develop AI/ML models to optimize business processes, design algorithms for data analysis and prediction, and integrate new technologies to enhance performance. Seeking to contribute to proof of value projects and support business development. EDUCATION

Third-year of Data Science Dec 2021 - Aug 2025

Van Lang University, Ho Chi Minh, Viet Nam GPA: 3.2/4.0 SKILLS

Programming Languages: Python, C#, SQL, R, Matlab

Tool and Frameworks: PyTorch, TensorFlow, Keras, Scikit-learn, Pandas, Numpy, Apache Spark, Matplotlib, Docker, Git. LANGUAGES

Vietnamese (Native proficiency) • English (Professional working proficiency) WORK EXPERIENCE

AI/ML Engineering Intern Feb 2024 - May 2024

Sprite Plus District 1, Ho Chi Minh City, Viet Nam Project: Detection and Classification of Lung Cancer using VGG16, Resnet50 Description: This project investigates the application of deep learning for the automatic detection and classification of lung cancer in medical images. It employs two powerful convolutional neural network (CNN) architectures: VGG16 and ResNet50. These pre-trained models leverage millions of labeled images to learn powerful features for image recognition. Duties: ·

Data preprocessing: Custom data of medical lung images (CT scans) and image processing such as extracting images to numpy array, reading the information, getting resolution from images, and collecting location and coordinates of cells cancer cells from the dataset. · Using a Transfer Learning model: VGG16 and ResNet50 models will be fine-tuned to detect and classify lung cancer with the final layers retrained on the lung cancer image dataset. · Model training and evaluation: Fine-tuned models will be trained on lung cancer image datasets to classify cancerous and healthy lungs. The model will be evaluated for performance through accuracy, precision, recall, and F1 score. Technology:

Deep Learning model: CNN, VGG16, Resnet50 ·

Frameworks: Pytorch, NumPy, Pandas, Scikit-learn, Keras, Matplotlib, PIL PROJECTS

Classification Object in CIFAR10 dataset using CNN model Apr 2024 - May 2024 Description:

Classification Object in CIFAR10 Dataset using CNN Model This repository contains the implementation of a Convolutional Neural Network

(CNN) model to classify images from the CIFAR-10 dataset. The CIFAR-10 dataset is a collection of 60,000 32x32 color images in 10 different classes, with 6,000 images per class. The classes include airplanes, cars, birds, cats, deer, dogs, frogs, horses, ships, and trucks. The dataset is divided into 50,000 training images and 10,000 test images. Duties:

Custom Dataset Creation: Developed and preprocessed a custom dataset from the CIFAR-10 dataset, including data normalization and augmentation techniques to enhance model generalization. Model Architecture Design: Designed and implemented a Convolutional Neural Network (CNN) architecture tailored to the CIFAR-10 dataset, optimizing layers and parameters for improved accuracy and performance. Model Training and Evaluation: Conducted extensive training and evaluation of the CNN model using TensorFlow and Keras, achieving a high accuracy rate on the test dataset and analyzing performance metrics such as accuracy, loss, and confusion matrix. Technology:

Deep Learning model: Design CNN architecture.

Frameworks: NumPy, Pandas, Scikit-learn, PyTorch, TensorFlow, Matplotlib, PIL, OpenCV Applications of Artificial Intelligence in ECG Signal Reconstruction Dec 2023 - Mar 2024 Description:

Heart disease is one of the leading causes of death globally. Early detection and diagnosis can help reduce mortality and improve treatment outcomes. Electrocardiogram (ECG) signals provide valuable information about the electrical activity of the heart. However, these signals can be corrupted by noise and artifacts, making it difficult to accurately interpret them. This project aims to reconstruct ECG signals, improve quality, and enable more accurate diagnoses.

Duties:

Collect ECG electrocardiogram data from the MIT-BIH Arrhythmia Database. Apply the Lowpass-Filter method to filter noise in the data set.

Observe the heart rate cycle to determine the data learning parameters of the LSTM model. Build LSTM model architecture, select optimize algorithms, and loss function to reconstruct ECG signals. Evaluate the model through the validation set, and test set to fine-tune the parameters. Technology:

Deep Learning model: RNN, LSTM.

Frameworks: NumPy, Pandas, Scikit-learn, Keras, TensorFlow, Matplotlib. Dataset: ECG datasets from MIT-BIH Arrhythmia Database. Weather Forecasting with Recurrent Neural Networks (RNNs) Oct 2022 - Dec 2022 Description:

This project explores the use of AI for weather forecasting in Basel City, Switzerland. Using deep learning techniques Recurrent Neural Networks (RNNs) to analyze historical weather data. The model utilizes Long Short-Term Memory (LSTM) units within the RNN architecture to capture long-term dependencies within the data. Duties:

Data collection and preprocessing: Gather historical weather data over many years including temperature, humidity, rainfall, wind direction, and many more labels. After collecting data, clean missing and abnormal values, and reformat variable types. Model selection and training: Choose suitable forecasting techniques RNN model. Train multiple models on historical data to evaluate and select the best-performing model.

Model monitoring: Track model accuracy in real-time as new data comes in. Technology:

Deep Learning model: RNN, LSTM

Frameworks: NumPy, Pandas, Scikit-learn, Keras, Matplotlib. Dataset: Weather Data from Meteoblue

CERTIFICATIONS

Rapid Minder - Machine Learning Professional

Rapid Minder - Data Engineering Professional

Rapid Minder - Applications & Use Cases Professional

Contact this candidate