Hi, I'm Saladin Zhalifunnas Ahfar.Building Across the Data Field

I work with data from analysis and database-backed workflows to machine-learning applications, turning it into clear insights and useful products.

About Me

My story

I am a Computer Science undergraduate at BINUS University in the Intelligent Systems stream. I am interested in the data field broadly, from organizing and analyzing data to translating findings into useful decisions and applications.

My projects include operational analytics and relational database workflows, predictive modeling, computer vision, and AI integration. They have helped me build a foundation across data analytics, data science, and applied AI while I continue developing my data engineering skills.

I value clean data, careful evaluation, clear communication, and collaborative problem-solving. I am looking for a data-focused internship where I can contribute and grow across analytics, data systems, or applied AI.

Based in

Tangerang, Indonesia

Open to data-focused internships in Data Analytics, Data Engineering, Data Science, and AI.

Education

Bachelor of Computer Science

BINUS University

Major
Computer Science
Streaming
Intelligent System
2024 — 2028GPA 3.73 / 4.00

Skills & Tools

Data Analytics & Systems

SQLPostgreSQLSupabaseGitHubPandasNumPyJupyterMatplotlibSeabornJoblib

AI / Machine Learning

PythonPyTorchTorchvisionTransfer LearningDeep LearningTensorFlowScikit-learnHugging FaceOpenCVXGBoostNatural Language ProcessingComputer Vision

Model & Product

FastAPIDockerStreamlitFlaskPillowReactViteJavaJava SwingObject-Oriented Programming

Full-Stack Engineering

Next.jsTypeScriptTailwind CSSPydanticSQLAlchemyAlembicPytestHTTPXESLint

Research

First page of Early Student Dropout Prediction Using Pre-Acceptance Variables with Gaussian Naive Bayes and XGBoost

Academic Research Project

Unpublished Manuscript

Early Student Dropout Prediction Using Pre-Acceptance Variables with Gaussian Naive Bayes and XGBoost

A co-authored study investigating whether pre-acceptance student information can support earlier, human-led dropout-risk screening.

The manuscript compares Gaussian Naive Bayes and XGBoost across two pre-acceptance feature settings using a cleaned sample of 3,630 student records, an 80:20 train-test split, and SMOTE handling. XGBoost achieved the strongest result in the Background-Application setting, with 0.7328 accuracy and a 0.6667 F1 score.

PythonScikit-learnXGBoostGaussian Naive BayesEducational Data Mining

EarlyDrop later evolved into a separate deployable application with a five-model benchmark and a threshold-tuned Random Forest workflow. This manuscript documents the original Gaussian Naive Bayes and XGBoost study.

Experience

Competitions

Let's connect

Let’s make data useful.

I am currently looking for internship opportunities where I can learn, contribute, and build useful technology with a thoughtful team.