WorkHub ๐
The Operating System for Modern Work. AI Managers + One-Stop Tooling on Atlassian Forge.
Inspiration ๐ก
500 times a day. Thatโs how often the average developer hits Cmd+Tab, switching contexts and bleeding focus.
2 in 3 people. Thatโs the ratio of employees endlessly repeating work across timezones because knowledge is siloed and invisible.
In the AI era, human resources shouldn't be wasted on context switching and redundant tasks.
We built WorkHub to stop the madness. We didn't just want a dashboard; we wanted an AI Manager that lives where you work, unifying your tools and automating the "work about work."
What it does ๐ ๏ธ
WorkHub is the one-stop destination for your entire workflow, embedded directly inside Jira.
- Unified Command Center: Kill the
Cmd+Tab. Access Figma, GitHub, Slack, Notion, and linear from a single glassmorphic interface without leaving Jira. - The AI Manager (Micro-tasks): WorkHub intelligently breaks down complex projects into bite-sized "Micro-tasks", automating the delegation and tracking that usually consumes an Engineering Manager's day.
- Smart Context: It eliminates timezone silos by surfacing essential updates ("Frequently Visited", "Bookmarks") so you don't repeat work done by your colleague in London 8 hours ago.
- Premium UX: A "Launchpad" experience that feels like a native OS, not a clunky plugin.
How we built it โ๏ธ
We pushed the Atlassian Forge platform to its absolute limits to bring this vision to life.
- Frontend: React + Vite for an instant, app-like feel.
- Design System: A custom Tailwind CSS implementation for that reliable, cohesive "OS" look.
- Intelligent Backend: Node.js running on Forge, orchestrating data between Jira and external tools.
- Data Layer: PostgreSQL with Drizzle ORM to manage the complex relationships between users, tools, and AI-generated micro-tasks.
Challenges we ran into ๐ง
- The "Sandbox" vs. "OS" Conflict: Building a tool that feels like an Operating System inside the secure sandbox of a Jira iframe (Custom UI) required creative engineering with CSPs and asset loading.
- Data Silos: unifying data from disparate sources (favicons, usage stats, task lists) to create a coherent "AI Manager" view without overwhelming the user.
- Performance: ensuring that a dashboard pulling potential real-time data loads instantly.
Accomplishments that we're proud of ๐
- Reducing the friction: We created an interface where a developer can theoretically spend their entire day without hitting
Cmd+Tabonce. - Visual Excellence: We proved that enterprise apps don't have to look boring. WorkHub looks and feels like a consumer product.
- Seamless Deployment: Automated the complex Forge deployment pipeline to get this live on Atlassian Online.
What we learned ๐
- The Cost of Context: Seeing the metrics on how fragmented work really is drove us to double down on the "One Stop" philosophy.
- Forge Potential: The platform is ready for "App-as-Infrastructure" solutions, not just small widgets.
What's next for WorkHub ๐ฎ
- Full AI Autonomy: Evolving the "AI Manager" from assigning tasks to verifying them (e.g., checking GitHub PRs automatically).
- Cross-Timezone Handoffs: AI-generated "End of Day" summaries that automatically brief the incoming team, killing the "2 in 3" redundancy stat for good.
- Integration Deep-Dive: Full read/write capabilities for connected tools (embedding Figma frames directly in the dashboard).
Built with โค๏ธ for Atlassian Codegeist.
Built With
- atlassian-jira
- mern
Log in or sign up for Devpost to join the conversation.