Data Engineer Intern - Python - SQL - AI/ML Enthusiast

Pham Bang

DATA ENGINEER INTERN

*********@*****.*** 094*-***-*** Ho Chi Minh City LinkedIn GitHub Objective

I aspire to become an Data Engineer to contribute to real-world projects, utilizing my knowledge and experience to deliver effective solutions for businesses. With my current skills, I am eager to continue learning and expanding my expertise in the future.

Technical Skills

Languages: Python, SQL

Computer Vision Tools: YOLO, OCR, OpenCV

AI Frameworks & Libraries: TensorFlow, PyTorch, Pandas, Numpy Machine Learning & Deep Learning: CNN, ResNet, VGG, U-Net, Transformer, ViLT, CLIP Others: Excel, Git, Docker, LangChain

Education

University of Economics Ho Chi Minh City — Major in Data Science 2022 – 2025 GPA: 3.7/4.0

Highlighted Courses: NLP (4.0), Machine Learning (3.5), Data Science (4.0), Statistics (4.0) Certificates:

Deep Learning Specialization – DeepLearning.AI, Coursera (Mar 2025) IBM Data Science Professional Certificate – IBM, Coursera (Aug 2024) IBM AI Engineering Professional Certificate – IBM, Coursera (Sep 2025) Experience

Metadata Solutions Co., Ltd. June - November 2025

Project: Clinic Appointment Scheduling & Patient Q&A Chatbot Role: AI Engineer / Data Engineer Built an automated ETL pipeline using Python to process 50+ DOCX files. Implemented a RAG system using vector embeddings and PostgreSQL to allow the chatbot to query patient data with 90% retrieval accuracy.

Participated in the design and development of the database architecture. Projects

1. Sarcasm Detection (UIT Data Science Challenge)

Developed a neural network that combines image and caption inputs to detect sarcasm in social media posts. Responsible for designing the model architecture, conducting experiments, and tuning performance metrics. 2. Animal Image Classification using ResNet

Fine-tuned pre-trained models (ResNet50 and VGG16) for classifying 132 animal species. Managed the entire pipeline from data preprocessing and augmentation to training and evaluating models. Also developed a simple interactive interface for real-time predictions.

3. Brain Tumor Segmentation using U-Net

Implemented the U-Net architecture to perform segmentation on medical images for brain tumor detection. Preprocessed image data, parsed annotations in COCO format, and trained the model on image-mask pairs. Achieved high accuracy (~95%) and documented the project thoroughly for reproducibility. in