Freelancers in DhakaFreelancers in Dhaka
UX Designer helping SaaS & startup teams turn user friction
1x
Hired
5.0
Rating
8
Followers
UX Designer helping SaaS & startup teams turn user friction
Analytics(1)
Power BI Data Analyst + ML AI Automation Expert
5.0
Rating
106
Followers
Power BI Data Analyst + ML AI Automation Expert
Cover image for End-to-End Machine Learning Pipeline for
End-to-End Machine Learning Pipeline for Telecom Customer Churn 1. The Business Problem Customer churn is a major challenge for telecommunications companies, driven by competition, service issues, and changing consumer preferences. This project was designed to transition the company from reactive support to proactive retention using data-driven strategies such as customer segmentation, personalized offers, and loyalty programs,. 2. Data Exploration & Insights (EDA) I performed a comprehensive descriptive analysis on a database of 7,043 customers with 21 distinct variables,. Key findings included: Contractual Risk: Customers on month-to-month contracts showed significantly higher churn compared to those on one- or two-year commitments,. Service Preference: While Fiber Optic plans were the most popular, they also represented a critical segment for monitoring due to their higher price points,. Financial Indicators: Churned customers had a higher average monthly charge of $74.44, compared to $61.27 for retained customers. Payment Behavior: The "Electronic Check" payment method was most strongly associated with service cancellation,. 3. Engineering & Preprocessing Pipeline To prepare the data for high-performance modeling, I implemented a rigorous preprocessing workflow: Data Cleaning: Removed irrelevant identifiers like customerID and addressed potential data quality issues. The dataset was verified to have zero missing or NaN values,. Feature Engineering: Applied Label Encoding to transform categorical text variables into a numerical format suitable for machine learning algorithms,. Data Splitting: Adopted a standard 80/20 train-test split to ensure the model could generalize effectively to unseen data,. 4. Model Development & Benchmarking I developed and benchmarked eight distinct machine learning algorithms to identify the most effective solution for this specific application: Linear & Probabilistic: Logistic Regression, Naive Bayes. Tree-Based: Decision Tree, Random Forest. Boosting Frameworks: AdaBoost, Gradient Boosting, XGBoost, and LightGBM,. 5. Performance Evaluation & Results Models were evaluated using ROC curves, confusion matrices, and detailed classification reports,. Winner: Logistic Regression achieved the highest accuracy at 81.83%,. Secondary Performers: Gradient Boosting (81.05%) and AdaBoost (80.98%) also showed strong predictive power. 6. Technical Conclusion This data-driven approach proves that proactive churn prediction is essential for business sustainability. By identifying that customers prioritize high-speed fiber optic services but are sensitive to pricing and contract terms, the company can now optimize its pricing and retention strategies to maximize user satisfaction and revenue.
4
712
Analytics(2)
UI/UX Designer | Product Designer | EdTech · FinTech · SaaS
New to Contra
UI/UX Designer | Product Designer | EdTech · FinTech · SaaS
Cover image for MagicFunnel - High-Converting SaaS Landing
MagicFunnel - High-Converting SaaS Landing Page Design The Business Challenge Most SaaS startups struggle with one major issue: their website looks good, but it doesn't convert visitors into paying customers. For MagicFunnel (an all-in-one marketing tool), the goal was to showcase a complex product simply, build immediate trust, and drive maximum sign-ups through strategic design. 💡 My Design Solution & Process Instead of just making a "pretty website," I engineered this landing page based on User Psychology and Conversion Rate Optimization (CRO) principles: Conversion-Driven Hero Section: Positioned a crystal-clear value proposition right at the top, supported by strong social proof and high-contrast primary CTA buttons to capture instant leads. Bento-Grid Feature Layout: Replaced boring text blocks with modern, interactive Bento cards to explain complex features (CRM, Analytics, Email Marketing) in under 5 seconds. Trust & Pricing Architecture: Designed clean, transparent pricing tiers alongside customer testimonials and active FAQs to eliminate buying friction. Developer-Ready Figma File: Built using strict Auto-layout, component properties, and clean naming conventions to save your developers hours of coding time. 📦 Deliverables Ready for Launch Full-length, responsive desktop and mobile landing page layouts. Complete visual style guide (Typography, active/hover states, color tokens). Production-ready assets and custom iconography. 🛠️ Core Expertise Demonstrated SaaS Website Design | UX Research | Landing Page Design | Figma UI/UX | Conversion Rate Optimization (CRO) | Startup Web Design | Responsive UI | Product Architecture Want to turn your website visitors into paying users? If you are a founder or product manager looking for a designer who understands business goals, let's talk. Click the "Inquire" or "Get in Touch" button on my profile, and let's discuss how we can scale your SaaS product.
2
67
Cover image for UI/UX Design for an All-in-One
UI/UX Design for an All-in-One Healthcare & Emergency Medical Assistant Mobile App Project Role: UI/UX Designer (Mobile App Design) Project Description / Overview This project focuses on designing a comprehensive MedTech & Telemedicine mobile application aimed at bridging the gap between patients, emergency medical services, and healthcare communities. The core objective was to centralize critical healthcare needs—such as finding blood donors, booking doctor appointments, and calling ambulances—into a single, stress-free user experience. The Challenge & The Solution The Challenge: During medical emergencies (like needing blood or an ambulance), users experience high stress. Most existing apps have cluttered interfaces, making it slow and confusing to find immediate help. The Solution: I designed a clean, high-contrast user interface with an intuitive grid system. By placing critical emergency services (Blood Donor, Ambulance, Hospital locator) right on the home screen with prominent Call-to-Action (CTA) buttons, I minimized user friction and reduced the time-to-action significantly. Key Deliverables & Features Designed Instant Blood Donor Module: Designed an easy-to-filter system showing donor availability tags (Available / Not Available) with direct calling options. Streamlined Doctor Booking: Created clear profile cards for popular doctors, displaying their expertise, ratings, and a seamless slot-booking journey. 24/7 Emergency Ambulance Banner: A high-visibility, one-tap booking section specifically tailored for urgent medical transit. Local Medical Services Grid: Clean icon-based navigation for nearby Hospitals, Diagnostic Centers, and Pharmacies. Community Trust Integration: Dedicated layout sections for social proof and community groups (Facebook/YouTube) to build user retention and credibility. Design & Tools Tools Used: Figma Design System: Modern Minimalist, accessible typography for all age groups, and an organic green accent color to evoke health, trust, and safety.
3
47
Cover image for Case Study: Transforming EdTech UX
Case Study: Transforming EdTech UX – Driving 40% More Conversions for TutorSheba Mobile App 📌 Executive Summary TutorSheba is a leading EdTech platform connecting students with verified home tutors. The goal was to redesign their core mobile experience to simplify tutor discovery, build instant user trust, and reduce booking friction for parents and students. Through intuitive visual hierarchy, smart categorization, and dynamic social proof, the new UI/UX design directly impacts user acquisition and conversion metrics. 🔍 The Challenge (The Problem Statement) Most EdTech apps fail because of choice paralysis and lack of trust. Parents looking for tutors are often overwhelmed by cluttered interfaces and skeptical about tutor credibility. TutorSheba's initial interface faced: High Bounce Rates: Users couldn't find their desired subjects quickly. Low User Trust: Premium and verified badge systems were not prominent enough. Friction in UX: The tutor search journey required too many taps, leading to drop-offs. 💡 The Strategy & Solution As a UX/UI designer focused on growth marketing, I restructured the layout based on Fitts's Law and Visual Hierarchy to guide user eyes toward high-converting zones. 1. High-Converting Hero Section The Psychology: Designed a prominent, value-driven banner ("1500+ Live Tuition Jobs Today") with a highly visible, thumb-accessible Call-to-Action (CTA) "Find Tutor Now". The Result: Creates instant FOMO (Fear Of Missing Out) and drives immediate action. 2. Frictionless Categorization (Browse by Subject) The Psychology: Implemented clean, color-coded grid icons for subjects (English, Bangla, Math, Physics). The Result: Minimizes cognitive load, allowing users to reach their specific needs in exactly one tap. 3. Trust Injected UI (Social Proof Integration) The Psychology: Placed a bold, high-contrast "Trusted & Verified" section with a micro-avatar stack representing "15k+ Verified Tutors." The Result: Directly combats parent skepticism, leveraging social proof right before they browse individual tutor profiles. 4. Verified Tutor Carousel The Psychology: Implemented an easy-to-scan horizontal card slider featuring "Premium" and "Verified" status tags directly on the tutor's profile cards. 🚀 The Business & Conversion Impact (UX Metrics) This design is not just beautiful; it is engineered for business growth: +42% Increase in Click-Through Rates (CTR) on the "Find Tutor Now" CTA. -35% Reduction in tutor search drop-offs due to optimized subject grids. Streamlined User Flow: Reduced the user journey from landing to booking by 3 steps. 🛠️ Toolstack & Keywords Tools: Figma, Adobe Illustrator, Prototyping
0
72
Cover image for Modern E-Commerce Landing Page Design
Modern E-Commerce Landing Page Design for Fashion & Lifestyle ✨ 📌 Project Overview In the highly competitive e-commerce space, visual appeal alone isn't enough—your website needs to convert visitors into buyers. This project focuses on a premium E-Commerce Landing Page designed to increase user engagement and drive sales through clean UI and intuitive UX. I created a minimalist, high-converting layout that reduces cognitive load, speeds up user decision-making, and provides a seamless shopping experience. 🛠️ Role & Deliverables My Role: UI/UX Designer (Research, Visual Design, Prototyping) Deliverables: High-fidelity Web UI, Responsive Layout Components, and Interactive Mockups. Timeline: 2 days Tools Used: Figma, Chatgpt, Claude, Stitch 💡 Key Features & Client Benefits High-Converting Hero Section: Designed a bold, structured hero header paired with a unified product showcase to capture user attention within the first 3 seconds. Optimized Navigation & Layout: Implemented a clean, left-sidebar category menu that streamlines product discovery and lowers bounce rates. Frictionless CTAs: Placed high-contrast "Shop Now" and "Explore Collections" buttons strategically to maximize the click-through rate (CTR). Smart Search Tags: Added popular search tags directly under the hero section, helping users find trending products faster and improving the overall user journey. 🎨 Visual & Design System Typography: Clean, geometric sans-serif fonts to ensure readability across all devices. Color Scheme: A sophisticated mix of deep charcoal, crisp white, and an emerald green accent to evoke a premium, eco-luxury feel. 🤝 Let's Work Together! Are you looking to redesign your e-commerce store or launch a high-converting landing page for your brand? I can help you create designs that don't just look beautiful, but also grow your business. 👉 Hire me directly on Contra to get started!
2
1
112
Analytics(1)
Visual designer who can code too.
1x
Hired
3
Followers
Visual designer who can code too.
Analytics(1)
Global Product Sourcing, Conversion Optimization, BI Report
5.0
Rating
Global Product Sourcing, Conversion Optimization, BI Report
Analytics(1)