Axis

Our Inspiration

Data centers generate hundreds of operational tickets daily, many with inconsistent priority levels, incomplete information, or unclear action items. We wanted to build a tool that combines AI intelligence with human oversight to streamline ticket management, improve accuracy, and reduce downtime.

The Purpose

Axis is a production-ready Streamlit application that:
    - Uses AI to auto-complete, validate, and enhance tickets.
    - Scores ticket priority objectively, overriding unrealistic user estimates.
    - Tracks tickets through a lifecycle:
            Draft → AI Review → Approved → In Progress → Completed
    - Maintains per-user activity logs and CSV export/import functionality.
    - Visualizes racks and simulates AR guidance for technicians.
    - Supports multi-user roles:
            Technician
            Engineer
            Admin
    - Integrates OpenRouter for AI-driven suggestions with intelligent fallback.

Building Process

We built our product into several phases where each stage stacks on one another.
    - Frontend & Backend:
            Streamlit, Python, and Plotly for interactive dashboards.
    - Data Storage:
            JSON files per user for tickets and audit logs.
    - AI Integration:
            OpenRouter API for priority scoring, validation, and automated enhancements.
    - Authentication:
            Simple role-based login with demo users.
    - UX Enhancements:
            Custom CSS for dark-mode, ticket cards, and metrics panels.

Technical Difficulties

While our team was working the product, we ran into problems that hindered our progress in developing the product.
    - Balancing AI suggestions with human decision-making, especially for priority conflicts.
    - Handling bulk CSV imports without blocking the UI or overwhelming the AI calls.
    - Ensuring consistent ticket lifecycle management across multiple user roles.
    - Designing intuitive visualizations for rack and server layouts with potential AR simulation.

Proud Accomplishments

Our team has come a long way in this hackathon, and we are excited about the achievements made throughout this process.
    - Fully functional multi-user system with per-user audit logs.
    - AI-powered ticket validation and priority scoring that can override human input intelligently.
    - Streamlined CSV bulk import/export workflows with real-time AI analysis.
    - Interactive dashboards showing ticket metrics, priority distributions, and tickets needing attention.

Learning Outcomes

During this hackathon, we have come to learn various things in order to improve our skills.
    - Combining AI with domain knowledge improves operational efficiency (requires careful UX design).
    - Real-time feedback and clear visualization of AI reasoning enhances trust and adoption by users.
    - Mocking AI during development greatly accelerates iteration without incurring API costs.

Future Vision

Although 24 hours is not much to work on a full-scale application, if we were to continue and keep developing on this product, we plan to implement these features:
    - Full AR-enabled rack visualization and guidance for technicians.
    - Integration with real-world ticketing systems like Jira or ServiceNow.
    - AI-driven auto-assignment and workflow suggestions for engineers.
    - Advanced analytics on ticket trends, root causes, and predictive maintenance.

Built With

• csv-import/export-support-authentication-&-security:-role-based-access-control
• multi-user-login-system-visualization-&-ar:-rack-visualization-via-plotly;-ar-guidance-simulation-(front-end-integration-ready)-devops-&-environment:-dotenv-for-environment-variables
• optional-jira-integration-for-enterprise-workflow-data-storage:-local-json-files-for-per-user-persistence
• pandas-(data-handling)
• pathlib
• plotly-(charts)
• programming-languages:-python-3.11+-frameworks-&-libraries:-streamlit-(ui)
• python
• requests-(api-calls)-cloud-&-apis:-openrouter-ai-(intelligent-ticket-analysis)
• uuid-for-unique-ticket-ids