Vu Quoc Hien

AI Engineer - FPT Software AI Center

Microsoft SharePoint RAG
Developed an Agentic RAG solution for question answering and document retrieval on Microsoft SharePoint using Azure Cloud Services, achieving a document hit rate exceeding 85% across over 100 SharePoint files and folders.

Document Q&A
Led the design and development of a robust data pipeline for analyzing complex PDF documents, resulting in a 40% reduction in data processing time through dynamic querying and insightful data extraction, and 92% overall accuracy on Q&A benchmark dataset, as evaluated by the RAGAS evaluation tool.

GenAI Chatbot
Developed an AI-powered chatbot utilizing advanced Retrieval-Augmented Generation techniques to enhance Toyota's Automated Bug Categorization Solution, improving model accuracy and operational efficiency by 30%.

Bioinformatics/AI Scientist - Genestory JSC

Cancer‑Risk Prediction
Developed a graph-based platform in Neo4j for gene–drug interaction data, enhancing the Pharmacogenomics (PGx) Database and achieving a fourfold increase in analysis speed and improved scalability.

Genetic AI Assistant
Led the development of an LLM‑RAG chatbot designed to boost internal consulting and customer service by integrating question answering, summarization, and information retrieval via a vector database.

SCARs -- Drug Response Prediction
Implemented a Severe Cutaneous Adverse Reactions (SCARs) project to mitigate patient risk by screening for harmful genes/variants in Allopurinol and Carbamazepine prescriptions using machine learning algorithms (Random Forest, SVM, and Gradient Boosting Trees), achieving accuracies of 96.9% and 81.2%, respectively.

AI Engineer - Vingroup Big Data Institute

Human Leukocyte Antigen (HLA) Imputation
Developed a neural network model incorporating fusion logic for HLA imputation in Vietnamese patient cohorts, achieving high accuracy (88–99%) and delivering significant clinical value.

Variant Annotation -- EML
Architected a 5 TB PostgreSQL database housing over 9 billion genetic variants and applied ensemble machine learning for comprehensive variant analysis.

AI Researcher - Modelling, Simulation, and Optimization Laboratory

• Research domains: Multi‑objective Optimization, Machine Learning, and Deep Reinforcement Learning.
• Main Projects:
  • Electric Vehicle Routing Problem: GitHub
  • Task Scheduling in Fog-Cloud Environments with Blockchain Integration: GitHub
• Side Projects:
  • Developed Monte Carlo Tree Search (MCTS) and deep reinforcement learning (DRL) methods for competitive board game agents: GitHub
  • Implemented a suite of DQN algorithms for the multi-agent, multiplayer Castle Invasion game: GitHub
  • Applied neuroevolution to build a high-performing Flappy Bird AI: GitHub