Nguyen Ba Long NBasLongz

Hi, I'm Nguyen Ba Long

Final-year Computer Science student at University of Information Technology - VNUHCM
Interested in focusing on Generative AI, RAG Systems, Computer Vision, and Big Data Pipelines
Building end-to-end AI workflows with Python, PyTorch, TensorFlow, LangChain, and FastAPI
Currently looking for an AI/ML Engineer Intern opportunity

Contact me

What I'm Up To

Current focus: AI/ML, Vietnamese NLP, RAG systems, and reinforcement learning
Recent projects: DashBot A3C Dashboard Generation, Temporal RAG for UIT Course Video Retrieval, Vietnamese Hate Speech Detector
Learning: MLOps, Docker, evaluation methods, and practical deep learning workflows
Open to: AI/ML Engineer Intern and NLP-related internship opportunities
Reach me: Email | LinkedIn

📌 Featured Projects

Generative AI, LLMs & RAG Systems

_{Tech: LangGraph, Gemini API, FastAPI, Qwen2-VL, Whisper, Milvus, Elasticsearch.}
_{Highlights: Engineered a directed state graph for multimodal video Q&A. Implemented hybrid retrieval (BM25 + Dense) fused via Reciprocal Rank Fusion (RRF) and validated using the RAGAS framework.}

_{Tech: Python, Flask, ChatGPT API, Q-Learning, Prompt Engineering.}
_{Highlights: Reproduced a research paper using LLMs to automatically generate and iteratively self-refine reward functions. Built a lightweight 2D Gridworld environment and a Flask API pipeline for automated code execution and feedback logging.}

Computer Vision & Multimodal AI

_{Tech: SigLIP2, Florence-2, YOLOv11, Milvus, Elasticsearch, Redis.}
_{Highlights: Processed 726 videos across 5 modalities. Built a hybrid search engine fusing Milvus and ES via Weighted RRF. Optimized precision with BGE Cross-Encoder and Rocchio Feedback, achieving top-1 scores > 0.93.}

_{Tech: TensorFlow, Keras, Scikit-learn, OpenCV, CNN-Transformer.}
_{Highlights: Designed a hybrid architecture with SE channel attention and MHSA bottleneck. Formulated an Edge-Aware Focal Tversky Loss to handle severe class imbalances, outperforming the baseline by 5.83% in Jaccard Index.}

Big Data, Data Mining & Analytics

_{Tech: Hadoop MapReduce, Apache Spark (RDD/DataFrame), PySpark, Apache Pig.}
_{Highlights: Developed large-scale data pipelines to process MovieLens and E-commerce datasets. Integrated real-time Computer Vision (YOLOv12) with PySpark and TCP Sockets for distributed batch video processing.}

_{Tech: Python, Scikit-learn, Underthesea, TF-IDF, LinearSVM.}
_{Highlights: Built a text preprocessing pipeline handling teencode and word segmentation. Trained a LinearSVM classifier achieving a Macro F1-Score of 0.6172 and 85.8% global accuracy on the UIT-ViHSD test set.}

Tech Stack & Tools

Languages

AI/ML & Data Science

NLP, RAG & Search

Backend & Databases

Frontend, Demo & Visualization

Tools & Platforms

## Detailed Analytics

Navigation Menu

Nguyen Ba Long NBasLongz

Organizations

Block or report NBasLongz