I began my career as a Project Control Engineer at a globally recognized EPC company, managing cost and schedules for large-scale infrastructure projects. Over time, I became deeply intrigued by the power of data and AI-driven systems, especially their potential to bring scalable solutions to the real world.

Now, I’m combining the structured mindset of engineering execution with hands-on machine learning development, from training deep learning models to deploying full-stack AI applications.

A hybrid model for dog breed classification and recommendation

PawMatchAI combines a CNN backbone with Transformer layers to create an advanced system for dog breed identification and matching. The project features several key innovations and achievements:

• Advanced Architecture: Hybrid model combining Convolutional Neural Networks with Transformer layers for enhanced feature extraction and classification accuracy
• Custom Feature Engineering: Proprietary Morphological Feature Extractor inspired by expert veterinary observation patterns
• High Performance: Achieved an impressive 88.7% F1-score on breed classification tasks
• Multi-functional Platform: Enables users to identify breeds, compare breed characteristics, and find optimal matches based on lifestyle preferences
• Recognition: Featured on Hugging Face's "Spaces of the Week" with over 32,000+ visits and 13,000 GPU runs
• Links: 🌐 Try the Demo | 🗂️ Explore the Project

Advanced multi-modal system for deep scene understanding

Vision Scout represents a sophisticated approach to visual intelligence, orchestrating multiple state-of-the-art models to transform complex visual data into comprehensive narratives:

• Multi-model Architecture: Seamlessly integrates YOLOv8, CLIP, Places365, and Llama 3.2 for comprehensive scene analysis
• Deep Understanding: Transforms raw visual data into human-readable stories and detailed scene descriptions
• Advanced Processing: Combines object detection, image classification, scene recognition, and natural language generation
• Community Recognition: Featured on Hugging Face's "Spaces of the Week" with significant user engagement
• Performance Metrics: Over 10,000 visits and 4,000+ GPU runs within three months, demonstrating strong user adoption
• Links: 🌐 Try the Demo | 🗂️ Explore the Project

Comprehensive data science and machine learning portfolio repository

The Learning Record repository chronicles a complete journey through data science fundamentals while tackling real-world business challenges. This collection represents hands-on experience solving complex problems across multiple industries and technical disciplines:

• Financial Analytics: Credit card fraud detection achieving 99% AUC performance using XGBoost and Bayesian optimization
• Credit Risk Assessment: Comprehensive credit score classification models with 85% accuracy utilizing ensemble methods and neural networks
• Natural Language Processing: Advanced MBTI personality prediction through sophisticated text analysis and NLP techniques
• Customer Analytics: E-commerce segmentation analysis using K-means and DBSCAN with optimized silhouette scores
• Time Series Forecasting: Retail sales prediction implementing ARIMA and SARIMAX statistical methodologies
• Signal Processing: Human activity recognition from smartphone sensor data using advanced preprocessing and dimensionality reduction techniques
• Audio Classification: Music genre classification from audio features with sophisticated feature engineering and model optimization

Links: 🗂️ Explore the Repository

I write about deep learning architectures, hybrid modeling, and AI system design, blending technical clarity with conceptual depth. All articles are selected as Deep Dives.

🔹 Machine Learning & AI Enthusiast
Hands-on in computer vision, NLP, and model deployment, with a focus on building useful, explainable, and well-integrated AI solutions.

🔹 Engineer Turned Data Explorer
From managing construction schedules and cost to training models, I carry the same structured, iterative mindset — whether it’s defining MVPs or analyzing feature contributions.

I also have experience with feature engineering, data preprocessing, and traditional machine learning pipelines. I believe in the principle of "Garbage in, garbage out": a lesson that applies as much to XGBoost as it does to a poorly defined CPM schedule. A well-prepared dataset, like a well-sequenced project timeline, determines everything downstream.

I’m open to new opportunities in:

👉 AI Product / Technical PM

👉 Machine Learning Engineer

👉 Data Scientist

If you're building AI products with real world impact, I’d love to collaborate.

"Every challenge is a puzzle — it's just waiting for the right combination of algorithms and insight."