Open to Data and AI Roles

Hi, I'm Ariful Mondal.

I turn messy enterprise numbers into interactive dashboards, automated pipelines and machine-learning models that guide real decisions.

Let's Talk Featured Work

pipeline.py

Data Analysis

Power BI

DAX

Excel

Power Query

Pivot Tables

SQL / MySQL

Data Cleaning

Dashboard Design

ServiceNow

Data Science

Python

Pandas

NumPy

Scikit-learn

Statistics

EDA

Feature Engineering

Regression

Classification

Random Forest

SVM

Model Evaluation

ML Pipelines

Matplotlib / Seaborn

Artificial Intelligence

Deep Learning

Neural Networks (ANN)

TensorFlow

Keras

Computer Vision

Transfer Learning

MobileNetV2

Offline AI Inference

Streamlit Apps

About

Analysis first, models second, always tied to a decision.

I am a data professional focused on converting messy enterprise numbers into interactive dashboards, automated intelligence, and clear operational strategies. My approach bridges deep mathematical analysis with real-world execution, helping organizations unlock key patterns and minimize revenue leakage.

With hands-on experience in supply chain and professional operations, I build complete descriptive pipelines, from raw business data to structured SQL databases to dynamic Power BI reports that track key metrics at a glance.

I extend those insights into data science and predictive analytics using Python, NumPy and Pandas to write automated clean-up logic and engineer high-performance machine learning models.

Power BI / Excel90%

SQL / MySQL82%

Python / Pandas85%

ML / Scikit-learn75%

Deep Learning / TensorFlow70%

Projects

Filter by discipline. Hover a card to inspect it.

NEW · FLAGSHIP PROJECT

CivicCura-AI

AI-Powered Offline Retinal Diagnostic Network for Primary Screening

// problem

Diabetic retinopathy leads to preventable blindness, but 80%+ of cases occur in low-resource settings lacking $10,000+ fundus cameras, stable internet, and specialized ophthalmologists.

// solution

An offline-first AI diagnostic system running locally on standard PCs. Paired with low-cost smartphone optical adapters (~$30), it delivers zero-latency primary screening, visual risk alerts, and downloadable PDF medical reports with no cloud or internet.

0of cases in low-resource settings

$30 vs $10k+adapter vs fundus camera

0internet connections needed

// how it works (click a step)

MobileNetV2APTOS 2019Python 3.13TensorFlowStreamlitReportLabOffline inferenceOne-click batch deploy

Interactive Lab: Data Runner

Dodge the noise, collect the signal. Hold jump for higher leaps, double jump mid-air, grab AI gems for a shield, press P to pause.

score 0 best 0 level 1

SPACE / TAP to start