Data Analyst & Business Intelligence Analyst

The studies focus on business related problems, such as factors affecting the working state of the workers and ultimately predicting employee attrition. In this work, Data Analytics study is applied in Human Resource Management (HRM) with the dataset which is the personnel record of a global company. The models employed were two machine learning classifiers; namely, Random Forest Classifier and Decision Tree Classifier for predicting employment status [17] and [22]. The overall accuracy of the Random Forest Classifier was 54% and feature significance was 95.8% for Salary_INR and F1-Score of the Active class was 0.70. The Decision Tree Classifier performed an overall accuracy of 69% with feature significance of Salary_INR as 61.0%, F1-Score as 0.81 for Active class. The findings offered some suggestions to the organisation to decrease employee turnover. View project →

I built this dashboard to answer a simple question: who is leaving, and has anything changed over ten years? It covers 100,000 employees, 10,010 of whom left, an overall attrition rate of 10.01%, with an average satisfaction score of 3.0/5 among those who resigned. - Trend: a combined area and line chart compares monthly active headcount with total attrition from January 2015 to January 2025, showing that the attrition pattern stayed consistent throughout the decade. - Department: Finance (10.5%), HR (10.3%) and Legal (10.2%) have the highest rates, while Engineering and IT are lowest at 9.6%. - Job title: Managers (10.26%) and Analysts (10.17%) lead, with Engineers lowest at 9.80%. - Education: employees with a Bachelor's degree make up nearly half (49.46%) of all attrition, followed by High School (30.46%). The main takeaway is that attrition is spread almost evenly across departments and roles. The gap between the highest and lowest is under one percentage point, so no single team or position is the problem. That points to company-wide factors rather than isolated ones. Department and job title filters let viewers explore each segment.

MR.COFFEE is a modern, warm, and inviting landing page designed for a specialty coffee shop brand based in Indonesia. Built as a Human-Computer Interaction (HCI) web prototype, the design combines intuitive navigation, rich visual storytelling, and clear call-to-action elements to deliver an engaging digital experience for coffee enthusiasts. Key Features & Design Highlights - Hero Section & Product Carousel: Features a bold, welcoming value proposition alongside an interactive "Top Products" highlight box with direct call-to-action ("Order Now") and smooth carousel indicators. - Intuitive Category Navigation: Dedicated visual category icons (Hot Coffee, Cold Coffee, Non-Coffee, Bread, Dessert) allow users to browse offerings effortlessly. - Brand Story & Leadership: Highlights the company's mission and background, featuring a dedicated profile section for the founder/CEO to build trust and connection with customers. -Integrated News & Updates: Incorporates a dynamic news and articles section with inspirational typography to keep visitors engaged with local and global updates. -Comprehensive Footer & Social Links: Includes organized quick links (Menu, About, Rewards) and social media integration for seamless navigation across all touchpoints. Technical & Design Specifications - Tool Used: Figma (UI/UX Design & Prototyping) - Target Domain: E-Commerce / Food & Beverage / Café Landing Page - Color Palette: Warm earth tones (coffee browns, cream whites, soft beige) designed to evoke comfort and warmth. - Typography: Modern sans-serif paired with bold headers for high readability and visual hierarchy.