MediAI V4
Research demo for ICU mortality risk prediction with a FastAPI service, LightGBM model, Redis caching, and LangChain-based retrieval. Not for clinical use.
Hello, I'm Uyen Huynh Khoi Minh
AI Engineer specializing in Biomedical ML & MLOps
Building research prototypes and engineering workflows that connect biotechnology, machine learning, and reproducible deployment.
Engineering Mindset
Bridging the gap between
Research & Deployment.
I focus on turning research models into reproducible demos and services with clear validation, automation, observability, and hardware-aware optimization.
CI/CD
ML API Automation
Benchmarks
Latency & Model Checks
๐ Education
Biotech Engineer
Nong Lam University Ho Chi Minh City
Expected 2026
๐ Languages
Vietnamese
Native
English
B2 Level
French
B2 Level
Chinese
HSK 4
๐ผ Experience
2+
Years AI Research
4+
Research Manuscripts
9+
Projects Completed
SYSTEM ARCHITECTURE
Key Systems & Solutions
Total: 02 Systems
T-DDI
Drug-Drug Interaction prediction research using molecular descriptors, Transformer-style tabular models, and class-imbalance-aware training. Public demo available.
THE STACK
Core Technologies & Tools
Python
PyTorch
TensorFlow
HuggingFace
Scikit-learn
XGBoost
LightGBM
Pandas
NumPy
LangChain
OpenAI API
FastAPI
Streamlit
Docker
Redis
PostgreSQL
Git
Linux
.NET
SQL
HTML/CSS
JavaScript
Projects & Research
Explore my portfolio of research projects that demonstrate the integration of AI and biotechnology, showcasing innovative solutions for healthcare, agriculture, and scientific discovery.
MediAI V4 ๐
ICU mortality risk prediction research demo with FastAPI, LightGBM, Redis caching, LangChain retrieval, Docker packaging, and CI/CD workflow. Intended for technical demonstration only, not clinical decision-making.
FastAPI
LightGBM
Redis
LangChain
Docker
Focus
MLOps Demo
T-DDI ๐
Drug-Drug Interaction prediction prototype using molecular descriptors, Transformer-style tabular modeling, focal loss, and a SMILES-to-feature pipeline. Results should be interpreted with the validation split and dataset context.
PyTorch
TabTransformer
QSAR
SMILES
Focus
DDI Prediction
BreastCNNs
Mammography image classification research prototype comparing deep learning models with a structured preprocessing pipeline. Grad-CAM is used for model interpretability, but the project is not validated for clinical diagnosis or screening decisions.
Python
Deep Learning
Breast
Medical AI
Mammography
Focus
Lesions Classification
Capybara
Product concept for a community learning platform built around peer teaching and structured feedback. The proposal explores semantic analysis, personalized learning paths, and in-depth feedback for language education.
Product
Full Stack
AI agent
Focus
Product
SAGE
Team project exploring a RAG-based healthcare information assistant with citations from sources such as PubMed, WHO, and CDC materials. This is a research and engineering prototype, not a compliant clinical system or medical advice tool.
AI agent
Full Stack
Healthcare
Focus
Healthcare
DeepGPT-DILI
Drug-induced liver injury prediction research prototype combining graph neural networks, XGBoost, Extra Trees, and attention-based sequence models. Public manuscript and repository links should be added before citing publication status or benchmark claims.
PyTorch
GCN
XGBoost
Attention-RNN
Docker
Focus
DILI Prediction
CRISPRON
Machine learning project for predicting CRISPR/Cas9 repair outcomes. The project should document genome build, coordinate conventions, sequence context, and validation data before biological conclusions are drawn.
Python
Machine Learning
CRISPR
Bioinformatics
Focus
Gene Editing
Energy Gel for Athletes
Food science project exploring energy gel formulations for endurance athletes, with attention to nutrition profile, bioavailability assumptions, and sustained energy release.
Food Science
Nutrition
Product Development
Research
Application
Sports Nutrition
Traffic Sign Detection
Computer vision project for traffic sign detection and classification, implemented with deep learning and OpenCV. The repository should document setup, sample outputs, and evaluation conditions before being presented as a production system.
Python
Computer Vision
CNN
OpenCV
Domain
Autonomous Systems
_d9750aa7-7903-4ed0-936e-70d01b22e08f-5ff882.png)
Cardiovascular Disease Prediction
Stacking ensemble learning experiment for cardiovascular disease prediction using multiple machine learning algorithms. Results require dataset, split, class balance, and baseline context before any clinical interpretation.
Python
Ensemble Learning
Healthcare AI
Stacking
Method
Ensemble Learning
Microbial Genome Analysis
Microbial genome analysis project covering genome assembly, annotation, and comparative genomics. Genome build, coordinate conventions, and sample metadata should be documented with the pipeline.
Bioinformatics
DNA Sequencing
Python
Genomics
Focus
Microbial Genome Analysis
Professional Journey
Bridging the gap between Biotechnology and Artificial Intelligence through research and leadership.
Startup
Co-founder
Capybara TechDirecting financial strategy and managing risk assessment for a growing tech venture.
- Direct financial strategy and planning
- Manage risk assessment and mitigation
- Lead strategic business development
Research
AI Researcher
Taipei Medical Univ.Developing GCN and Transformer experiments for biomedical machine learning.
- Developed DILI prediction experiments with GNN and ensemble models
- Built T-DDI modeling workflows for molecular interaction prediction
- Peer Reviewer for PLOS ONE
Applied AI
AI Researcher
AI VietnamFocused on LLM Research (VLAI Team) and mathematical reasoning efficiency.
- Co-authored Multi-Mode CoT research (2025)
- Fine-tuned Llama 3.2 3B (80.06% GSM8K)
- Built automated Data Curation Pipeline
Internship
Software Dev Intern
Mat Bao Corp.Developed and maintained web applications ensuring optimal performance.
- Backend development with .NET framework
- SQL database management and optimization
Get In Touch
I'm always interested in discussing new opportunities, collaborations, or projects that combine biotechnology with artificial intelligence. Let's connect and explore how we can work together!
Contact Information
Phone
+84 397 849 278
Location
Ho Chi Minh City, Vietnam
University
University of Agriculture and Forestry HCMC
Available for opportunities
Open to internships, research collaborations, and full-time
positions in biotechnology and AI.