Skip to content

README.md

Latest commit

 

History

History
188 lines (122 loc) · 6.14 KB

README.md

File metadata and controls

188 lines (122 loc) · 6.14 KB
Raw

VibeWork — AI Workplace Hub

Offline-First AI-Powered Workplace Hub. All AI runs locally on your device.

Project info

VibeWork is a modern, privacy-focused workplace hub that leverages on-device AI for tasks, knowledge management, and productivity assistance.

Detailed Application Description

VibeWork is a role-aware workplace operations platform designed for teams that need productivity tooling, status visibility, and AI assistance without relying on cloud inference. The application combines task tracking, daily work logging, team communication, project risk monitoring, onboarding support, and skill growth insights in one interface.

The core product idea is simple: every common team workflow should be available offline first, with AI features enhanced locally when a model is available and graceful fallbacks when it is not.

Product Goals

  • Provide a single workspace for day-to-day team coordination.
  • Keep sensitive organizational context local by default.
  • Offer role-specific experiences instead of one generic dashboard.
  • Support intermittent connectivity and local persistence.
  • Make AI assistive, contextual, and optional rather than blocking.

How the App Works

When users sign in, they choose a role and enter the workspace with a role-filtered navigation sidebar. The app seeds local demo data on first launch, then stores all changes in browser storage so users can continue operating without backend setup.

Most pages follow this pattern:

  1. Load data from local storage tables.
  2. Render role-appropriate widgets and actions.
  3. Trigger AI helper actions (summaries, analysis, suggestions).
  4. Persist updates locally and refresh UI state.

Core Modules

1. Dashboard

  • Daily briefing and productivity snapshot.
  • Task completion summary and pending workload.
  • Quick links to common workflows.
  • Recent activity feed.

2. Task Management

  • Kanban-style task flow (backlog, in-progress, review, done).
  • Priority filtering and task creation.
  • AI prioritization suggestions for ordering work.

3. Daily Logger

  • Daily work notes with date-scoped entries.
  • AI-generated summaries from raw text.
  • Mood/sentiment indicators for personal trend awareness.
  • Voice input support for fast logging.

4. Knowledge Vault

  • Folder-based internal knowledge organization.
  • Search and contextual retrieval helpers.
  • AI-assisted SOP/procedure generation.

5. Announcements

  • Organization-wide update publishing.
  • Urgency and pinned messaging support.
  • Text-to-speech playback for accessibility and field use.

6. Code Review

  • AI-assisted review summaries and issue suggestions.
  • Developer-focused quality checks and guidance output.

7. Burnout Detection

  • Worklog sentiment trend tracking.
  • AI risk commentary for early intervention decisions.

8. Risk Predictor

  • Project risk score display.
  • AI-generated risk reasoning and recommendation output.

9. CI/CD Tracker

  • Build status timeline (success, failed, running).
  • Duration and commit context visibility.
  • AI failure analysis helper for quick triage.

10. Onboarding Bot

  • Guided onboarding progress tracking.
  • Step-based completion and quiz insights.

11. Skill Tracker

  • Category-wise competency tracking.
  • Radar-chart visualization for strengths and gaps.
  • AI recommendations for growth priorities.

12. Settings

  • Theme and session controls.
  • Local data export and reset.
  • On-device model lifecycle management (download/load).

Role-Based Experience

The app supports five primary roles:

  • employee
  • team-lead
  • manager
  • hr-admin
  • field-worker

Navigation and page access are role-filtered so each persona sees relevant workflows (for example, managers get risk and burnout views, HR gets announcement and onboarding control, field workers get voice-first paths).

AI Behavior and Offline Strategy

VibeWork uses RunAnywhere for on-device LLM inference. If the local model is not available, AI features degrade gracefully to deterministic mock responses so users can continue exploring and working without hard failures.

This gives a practical two-tier AI strategy:

  • Tier 1 (preferred): On-device model loaded and used for local inference.
  • Tier 2 (fallback): Mock AI responses preserve usability and demos.

Data and Persistence Model

Application data is stored locally per collection (users, tasks, logs, announcements, knowledge items, skills, projects, onboarding progress, CI builds). A lightweight local DB abstraction provides CRUD operations and seeded sample records for immediate usability.

Session and UI preferences (role, user identity, theme, model selection, sidebar state) are persisted through Zustand middleware.

UI/UX Characteristics

  • Responsive layout with desktop sidebar and mobile overlay navigation.
  • Glassmorphism-inspired visual style with smooth transitions.
  • Small, focused cards and modules to reduce cognitive load.
  • Clear status signaling for network, AI readiness, and model state.

Typical End-to-End Flow

  1. User signs in and selects role.
  2. Dashboard shows current workload and briefing.
  3. User updates tasks and submits daily log.
  4. AI summarizes work and suggests priorities or risks.
  5. Team/management views reflect updated operational context.

This makes VibeWork suitable as a hackathon prototype, an internal proof-of-concept for private AI operations tooling, or a base for a production-ready workplace assistant platform.

Getting Started

Prerequisites

  • Node.js & npm installed

Installation

# Step 1: Clone the repository
git clone <YOUR_GIT_URL>

# Step 2: Navigate to the project directory
cd <YOUR_PROJECT_NAME>

# Step 3: Install the necessary dependencies
npm install

Development

# Start the development server
npm run dev

Technologies Used

  • Framework: React with Vite
  • Language: TypeScript
  • Styling: Tailwind CSS & shadcn/ui
  • AI Engine: Runanywhere SDK (On-device LLM)
  • Database: Local browser storage abstraction
  • State Management: Zustand
  • Icons: Lucide React

Powered by Runanywhere

This project uses Runanywhere for on-device AI inference, ensuring privacy and offline functionality.