✦ Open to data science & ML research opportunities — Let's connect →
Data Scientist & ML Engineer

Turning Data Into
Intelligence

Passionate about building machine learning pipelines, exploring NLP and computer vision, and solving real-world problems with statistical rigor and deep learning.

View My Projects Download Resume
GH LI KG EM
About Me

Who I Am

I'm a data scientist and machine learning engineer with a strong foundation in statistics, probability, and experimental design. I believe that good ML starts with understanding the data — and I'm equally comfortable building regression models in R, training YOLO for computer vision, or experimenting with transformer architectures for NLP.

My interests span the full ML spectrum: from feature selection and hyperparameter tuning to cutting-edge retrieval-augmented generation (RAG) and vector databases. I'm always reading research papers and looking for ways to turn theory into practice.

6+
ML Domains
3
Languages
R
Statistical Tool
RAG
Current Focus
Expertise

Skills & Domains

Six core areas of my ML skill set — from statistics to deep learning.

Statistics

Probability Theory Statistical Methods Regression Analysis Feature Selection Experimental Design Multi-Linear Regression R Programming

Coding

Python Java OOP Principles Database Concepts R SQL

NLP

NLP Pipelines Transformers RAG Vector Databases Research Paper Reading Tokenization & Embeddings

Computer Vision

Multi-input Analysis YOLO CLIP Image Processing Audio Analysis Tabular + Vision Fusion

Machine Learning

ML Pipelines Feature Engineering Hyperparameter Tuning Model Evaluation Cross-Validation

Deep Learning

Neural Networks Backpropagation CNN Architectures Transfer Learning Loss Functions
Featured Work

Projects I've Built

Course projects and personal work spanning statistics, computer vision, and NLP.

Population Age Prediction

Built a multi-linear regression model in R to predict population age distributions. Applied feature selection techniques and validated using statistical methods including ANOVA and residual analysis.

R Regression Feature Selection Statistics
View on GitHub → Report →

Multi-Input Analysis (Image + Audio + Tabular)

Developed a multi-modal system combining YOLO for object detection and CLIP for image-text alignment. Integrated audio and tabular data streams for comprehensive analysis.

YOLO CLIP Python CV
View on GitHub → Demo →
■

NLP Pipeline

End-to-end NLP pipeline with tokenization, embedding, transformer models, and RAG integration using vector databases.

NLP Transformers RAG
●

ML Pipeline Framework

Modular ML pipeline with automated feature selection, hyperparameter tuning, cross-validation, and model evaluation.

Python Scikit-learn MLOps
▲

Neural Network From Scratch

Implemented neural network fundamentals including backpropagation, activation functions, and loss optimization from the ground up.

Deep Learning NumPy Math
Education

Coursework & Training

Completed coursework in statistics, probability, and machine learning methods.

Probability & Statistics

Core probability theory, distributions, hypothesis testing, confidence intervals, and Bayesian inference.

Statistical Methods

Advanced statistical modeling including regression, ANOVA, experimental design, and feature selection techniques.

Experimental Design

Design of experiments, A/B testing methodology, confounding variables, and causal inference.

Object-Oriented Programming

OOP principles in Python and Java — encapsulation, inheritance, polymorphism, and design patterns.

Database Concepts

Relational database design, SQL, normalization, indexing, and basic query optimization.

Multi-Linear Regression (R)

Course project: predicted population age using MLR in R with feature selection and model diagnostics.

Research Interests

What I'm Exploring

RAG & Vector Databases

Building retrieval-augmented generation pipelines with vector embeddings, dense retrieval, and knowledge base integration.

Transformer Models

Deep understanding of attention mechanisms, BERT/GPT architectures, fine-tuning strategies, and model interpretability.

Multi-Modal Learning

Combining vision, language, and audio — inspired by CLIP and YOLO — for richer representation learning.

Read My Notes Let's Collaborate →
Get in Touch

Let's Work Together

Have a project, research idea, or just want to chat about ML? Drop me a message.

Send Message