Skip to main content

Dashboard Overview

The Alactic AGI dashboard is your central interface for managing documents, scraping websites, analyzing content, and monitoring processing jobs. This guide provides a complete walkthrough of the dashboard interface and key features.

Accessing the Dashboard​

After deployment, access your dashboard at:

https://<your-vm-ip-address>

Or use the custom URL from deployment outputs:

https://alactic-<unique-id>.<region>.cloudapp.azure.com

Login with Deployment Key: Enter your deployment access key (format: ak-xxxxx) on the login page.

Dashboard Layout​

The dashboard uses a clean, modern interface with these main sections:

Top Navigation Bar​

Left Side:

  • Alactic AGI Logo - Click to return to home
  • Version Number - Shows current version (e.g., v3.1.0)

Right Side:

  • Dashboard - Main processing interface
  • Settings - Configuration and key management
  • Help - Documentation and support
  • Logout - End session

Main Content Area​

The central workspace where you process documents and view results.

Key Features:

  • Upload PDFs or enter URLs
  • Configure processing options
  • View extracted content
  • Download results

Status Panel (Right Sidebar)​

Real-time information about your deployment:

  • Plan Type - Current plan (Free/Pro/Pro+/Enterprise)
  • Usage Statistics - Documents processed, storage used
  • Model Information - Available AI models
  • Processing Queue - Active jobs count

Home Page Features​

Document Processing Tabs​

1. PDF Upload Tab

Upload and process PDF documents:

Features:

  • Drag-and-drop file upload
  • Multi-file selection (batch upload)
  • File size validation
  • Processing progress bar

Plan Limits:

  • Free: 20 PDFs max
  • Pro: 100 PDFs max
  • Pro+: 500 PDFs max
  • Enterprise: Unlimited

Supported:

  • PDF format only
  • Max 50MB per file (Free)
  • Max 100MB per file (Pro/Pro+)
  • Max 500MB per file (Enterprise)

2. URL Scraping Tab

Extract content from websites:

Features:

  • Single URL entry
  • Bulk URL processing (paste multiple URLs)
  • URL validation
  • Site detection (detects if site blocks scraping)

Plan Limits:

  • Free: 50 URLs max
  • Pro: 200 URLs max
  • Pro+: 1,000 URLs max
  • Enterprise: Unlimited

URL Formats Supported:

  • HTTP and HTTPS
  • Authentication (if provided)
  • Redirects handled automatically

3. Batch Processing Tab

Process multiple documents or URLs at once:

Features:

  • Upload multiple PDFs simultaneously
  • Paste list of URLs (one per line)
  • Queue management
  • Progress tracking for each item
  • Bulk download results

Best For:

  • Research projects requiring 50+ documents
  • Competitive analysis (scrape 20+ competitor sites)
  • Content aggregation workflows
  • Regular scheduled processing

Processing Options​

Before submitting documents or URLs, configure these options:

Model Selection​

Alactic GPT-4o:

  • Full-featured model
  • Best for complex analysis
  • Higher cost ($2.50/1M tokens)
  • Available on Pro, Pro+, Enterprise plans

Alactic GPT-4o mini:

  • Cost-effective model
  • Good for straightforward extraction
  • Lower cost ($0.150/1M tokens)
  • Available on all plans

Recommendation: Use GPT-4o mini for simple extraction, GPT-4o for complex reasoning.

Analysis Depth​

Quick Extract:

  • Fast processing (5-10 seconds per document)
  • Basic text extraction
  • Minimal analysis
  • Good for data collection

Standard Analysis:

  • Balanced processing (15-30 seconds per document)
  • Text extraction + summarization
  • Key points identified
  • Default option

Deep Analysis:

  • Comprehensive processing (45-90 seconds per document)
  • Full content understanding
  • Detailed insights
  • Entity extraction
  • Sentiment analysis
  • Best for research and decision-making

Output Format​

JSON:

  • Structured data format
  • Easy to parse programmatically
  • Includes metadata
  • Best for API integration

Markdown:

  • Human-readable format
  • Preserves formatting
  • Easy to review
  • Best for documentation

Plain Text:

  • Simple text extraction
  • No formatting
  • Smallest file size
  • Best for basic content retrieval

Additional Options​

Vector Storage:

  • ☑ Enable vector embeddings for semantic search
  • Allows similarity queries across documents
  • Slightly increases processing time (+2-5 seconds)
  • Required for "Find Similar" feature

Content Chunking:

  • Breaks large documents into smaller sections
  • Better for long PDFs (50+ pages)
  • Improves processing speed
  • Enables section-level analysis

Results View​

After processing completes, view results in several formats:

Card View (Default)​

Each processed document appears as a card showing:

  • Document Name/URL - Source identifier
  • Processing Status - Complete, Processing, Failed
  • Model Used - GPT-4o or GPT-4o mini
  • Processing Time - Seconds taken
  • Token Usage - Input/output tokens consumed
  • Cost - Calculated cost for this document

Actions:

  • View - See full extracted content
  • Download - Save results locally
  • Reprocess - Run again with different options
  • Delete - Remove from results

List View​

Compact table view showing:

  • Document name
  • Status
  • Date processed
  • Model
  • Cost

Best for: Reviewing large batches (50+ documents)

Full Content View​

Click "View" on any document to see:

Summary Section:

  • Auto-generated summary (150-300 words)
  • Key topics identified
  • Main conclusions

Extracted Content:

  • Full text from document
  • Preserves structure (headers, lists, tables)
  • Formatted for readability

Metadata:

  • File size
  • Page count (PDFs)
  • Word count
  • Language detected
  • Processing timestamp

Entities (if Deep Analysis enabled):

  • People names
  • Organizations
  • Locations
  • Dates
  • Financial figures

Sentiment Analysis (if Deep Analysis enabled):

  • Overall sentiment: Positive/Neutral/Negative
  • Confidence score
  • Sentiment distribution by section

Bulk Actions​

Select multiple documents to:

  • Download All - ZIP file with all results
  • Reprocess All - Batch reprocessing with new options
  • Delete All - Bulk deletion
  • Export to CSV - Metadata table

Search and Filter​

Search processed documents by:

  • Keyword - Full-text search across all content
  • Title - Search document names
  • Date Range - Filter by processing date
  • Model - Filter by GPT-4o or GPT-4o mini
  • Status - Filter by Complete/Failed/Processing

Semantic Search (Pro+/Enterprise Only)​

Find documents by meaning, not just keywords:

Example:

  • Query: "climate change impacts"
  • Finds documents about "global warming effects" and "environmental consequences"
  • Uses vector embeddings for similarity matching

How to Use:

  1. Click "Semantic Search" tab
  2. Enter natural language query
  3. View results ranked by relevance
  4. Click to view full document

Settings Page​

Access via top navigation → Settings

Deployment Information​

Deployment Key:

  • View your access key (partially masked: ak-abc...xyz)
  • Copy to clipboard
  • Cannot be changed - requires redeployment

Deployment ID:

  • Unique identifier for your instance
  • Used for support requests

Azure Region:

  • Shows where deployment is hosted
  • Cannot be changed without redeployment

Usage Statistics​

Real-time metrics:

Document Processing:

  • PDFs processed this month
  • URLs scraped this month
  • Remaining quota (Free/Pro/Pro+ plans)

Storage:

  • Current storage used
  • Storage limit
  • Percentage utilized

Token Usage:

  • Input tokens consumed (current month)
  • Output tokens consumed (current month)
  • Estimated cost
  • Cost breakdown by model

Cost Tracking:

  • Infrastructure costs (VM, storage, Cosmos DB)
  • Model API costs (per 1M tokens)
  • Total monthly spend estimate

Model Configuration​

Available Models: Shows which models are available on your plan:

  • Alactic GPT-4o: Available / Not Available
  • Alactic GPT-4o mini: Available

Default Model: Select which model to use by default for new processing jobs.

Temperature Setting: Adjust model creativity (0.0 - 1.0):

  • 0.0: Deterministic, consistent output
  • 0.5: Balanced (default)
  • 1.0: Creative, varied output

Notification Preferences​

Email Notifications:

  • Processing job completed
  • Processing job failed
  • Daily usage summary
  • Monthly cost report

Webhook Notifications (Enterprise Only): Configure webhook URL to receive processing events in real-time.

API Access​

API Endpoint:

https://<your-vm-ip>/api/v1

Authentication: Use your deployment key in API requests:

curl -H "X-Deployment-Key: ak-xxxxx" \
     https://<your-vm-ip>/api/v1/process

API Documentation: Link to full API reference with examples.

Security Settings​

IP Whitelisting (Pro+/Enterprise): Restrict dashboard access to specific IP addresses:

  • Add allowed IPs (one per line)
  • Blocks all other IPs
  • Good for corporate deployments

Session Timeout: Set automatic logout time:

  • 30 minutes (default)
  • 1 hour
  • 4 hours
  • Never (not recommended)

Help and Support​

In-Dashboard Help​

Help Icon (?): Hover over any field to see contextual help tooltip.

Quick Start Guide: Step-by-step walkthrough for first-time users.

Video Tutorials: Links to video guides for common tasks.

Documentation​

Access Full Documentation: Click "Help" → "Documentation" to visit this documentation site.

Sections:

  • Getting Started
  • Processing Documents
  • API Reference
  • Troubleshooting

Support Channels​

Email Support: support@alacticai.com (Enterprise plans: 24-hour response)

GitHub Issues: Report bugs and request features: alactic.io/support

Community Forum: Get help from other users: community.alactic.ai

Keyboard Shortcuts​

Speed up your workflow with these shortcuts:

ShortcutAction
Ctrl+UOpen PDF upload
Ctrl+LFocus URL input
Ctrl+EnterSubmit processing job
Ctrl+FSearch documents
Ctrl+SGo to Settings
Ctrl+HOpen Help
EscClose modal/cancel

Mobile Interface​

The dashboard is responsive and works on mobile devices:

Supported:

  • Tablets (iPad, Android tablets)
  • Smartphones (iOS, Android)

Limitations on Mobile:

  • Batch processing harder (small screen)
  • Some charts not visible
  • Recommend desktop for production use

Dashboard Performance​

Loading Times:

  • Initial load: 1-2 seconds
  • Processing submission: <500ms
  • Results refresh: <1 second

Browser Requirements:

  • Modern browser (Chrome, Firefox, Safari, Edge)
  • JavaScript enabled
  • Cookies enabled (for session management)

Recommended Browsers:

  • Chrome 90+
  • Firefox 88+
  • Safari 14+
  • Edge 90+

Next Steps​

Now that you understand the dashboard:

  1. Process Your First Document
  2. Set Up Batch Processing
  3. Configure API Access
  4. Monitor Usage

Troubleshooting​

Dashboard Won't Load:

  • Check VM is running
  • Verify IP address is correct
  • Check firewall rules allow port 443

Can't Login:

  • Verify deployment key is correct
  • Check key hasn't expired
  • Try different browser

Processing Stuck:

  • Check service status in Settings
  • Verify model deployment is healthy
  • Check logs for errors

For detailed troubleshooting, see Common Issues.