A production‑grade frontend application for a home assistant robot platform, designed with clarity, reliability, and human‑centric UX at its core.
This repository focuses on the frontend system — the user interface layer that connects people to a real, live robotic and healthcare‑oriented platform.
The application is fully deployed and can be explored live.
🌐 Live Demo: https://home-buddy-v2.vercel.app/
🧠 Backend API: https://github.com/azedta/home-buddy-v2
HomeBuddy’s frontend is designed around three real user contexts:
- 👵 Elderly users who need calm visuals, large touch targets, and confidence in every action
- 👨⚕️ Caregivers & administrators who need visibility, control, and predictable workflows
- 🤖 A physical robot system, where UI decisions must reflect real‑world behavior
Every screen, component, and interaction is intentional and purpose‑driven.
- Calm interfaces → soft color palettes, low visual noise, stable layouts
- Predictable interactions → consistent spacing, familiar patterns, no surprises
- Accessibility‑aware UI → readable typography, forgiving interactions, clear feedback
- Human‑centered language → friendly wording over technical jargon
The goal is not visual flash — it is clarity, trust, and usability.
Screenshots below demonstrate real application flows and interaction patterns.
- Clean, reassuring authentication flows
- Real‑time validation with clear feedback
- Password strength guidance during signup
- Session‑aware and persistent login handling
- Calm success and error messaging
A core part of the application’s frontend experience.
- Clear separation between Elderly and Admin/Caregiver views
- UI structured around real medication routines
- Emphasis on readability and reassurance
- Card‑based layouts designed to reduce cognitive load
- Day‑based planners instead of dense timelines
- Window‑based schedules for better comprehension
- Visual rhythm (morning / afternoon / night)
- Status changes that are deliberate and explicit
- Status cards designed to inform without alarming
- Clear indicators for battery, tray, docking, and dispenser state
- Visual metaphors instead of raw technical data
- Companion‑like presentation rather than machine‑centric UI
A complete frontend notification inbox, not a simple alert system.
- Severity‑aware visuals (critical, warning, info, success)
- Clear, human‑friendly titles and messages
- Admin view across multiple users
- Filtering, pagination, and search
- Expandable details with optional technical transparency
This frontend is fully integrated with a live backend system:
- All UI actions reflect real backend state
- Authentication, medication schedules, notifications, and robot status are live
- Role‑aware rendering matches backend authorization logic
- UI state updates accurately reflect system events
The application is deployed end‑to‑end, allowing the entire system to be explored in real time.
- ⚛️ React (functional components, hooks)
- ⚡ Vite (modern dev server & build tool)
- 🎨 Tailwind CSS (utility-first, design-system driven UI)
- 🧭 React Router (multi-role routing & protected flows)
- 🧠 Zustand (authentication & global state management)
- 🔌 Fetch-based API layer (
apiFetchabstraction) - 🔔 react-hot-toast (non-intrusive feedback & alerts)
- 🧩 lucide-react (consistent icon system)
- Component‑driven UI design
- Predictable state management
- Role‑aware rendering within a single application
- Consistent visual language across all modules
- No placeholder or decorative UI — every component is functional
This frontend represents a complete, production‑ready user interface for a complex, real‑world system involving healthcare routines, robotics, and multi‑role users.
The focus throughout the application is clarity, stability, and trust — principles reflected in both the UI design and the frontend architecture.
This project is proprietary and protected under an All Rights Reserved license.
The source code is provided for viewing and evaluation purposes only as part of a personal portfolio. Any use, reproduction, modification, or distribution without explicit permission from the author is prohibited.