Analytics & Monitoring

Analytics & Monitoring

UniCraft provides comprehensive analytics and monitoring capabilities to help you understand and optimize your AI usage patterns.

Dashboard Overview

The analytics dashboard provides real-time insights into your AI usage:

Key Metrics

• Total Requests: Number of API calls made
• Total Cost: Cumulative spending across all providers
• Average Response Time: Mean response time across all requests
• Success Rate: Percentage of successful requests
• Active Models: Currently used AI models

Time-based Analysis

• Hourly, daily, weekly, and monthly views
• Trend analysis and forecasting
• Seasonal pattern detection

Usage Analytics

Request Patterns

Analyze your request patterns to optimize usage:

const analytics = await unicraft.analytics.getUsage({
  start_date: "2024-01-01",
  end_date: "2024-01-31",
  group_by: "day",
  metrics: ["requests", "cost", "tokens"],
});

Provider Comparison

Compare performance across different providers:

• Response times
• Cost per request
• Success rates
• Model availability

Model Performance

Track performance metrics for each model:

• Average response time
• Token usage
• Cost efficiency
• Quality scores

Cost Analytics

Spending Trends

Monitor your spending patterns:

• Daily, weekly, monthly cost trends
• Cost per request analysis
• Provider cost comparison
• Budget vs. actual spending

Cost Optimization Insights

Get recommendations for cost optimization:

• Underutilized models
• Expensive request patterns
• Alternative model suggestions
• Batch processing opportunities

Budget Management

Set and monitor spending budgets:

const budget = await unicraft.budgets.create({
  name: "Monthly AI Budget",
  amount: 1000,
  period: "monthly",
  alerts: [0.8, 0.9, 1.0], // Alert at 80%, 90%, and 100%
});

Performance Monitoring

Response Time Tracking

Monitor response times across providers and models:

• Average response time
• 95th percentile response time
• Response time distribution
• Performance trends

Error Rate Monitoring

Track and analyze errors:

• Error rate by provider
• Error type classification
• Error trend analysis
• Root cause analysis

Availability Monitoring

Monitor provider availability:

• Uptime statistics
• Service health checks
• Outage detection
• Recovery time tracking

Custom Metrics

Custom Dashboards

Create custom dashboards for specific use cases:

const dashboard = await unicraft.dashboards.create({
  name: "Production Monitoring",
  widgets: [
    {
      type: "line_chart",
      metric: "response_time",
      provider: "openai",
    },
    {
      type: "gauge",
      metric: "success_rate",
      threshold: 0.95,
    },
  ],
});

Alerts and Notifications

Set up custom alerts for important metrics:

const alert = await unicraft.alerts.create({
  name: "High Error Rate",
  condition: "error_rate > 0.05",
  duration: "5m",
  channels: ["email", "slack"],
});

Reporting

Automated Reports

Schedule automated reports:

• Daily usage summaries
• Weekly cost reports
• Monthly performance reviews
• Custom report templates

Export Options

Export data in various formats:

• CSV for spreadsheet analysis
• JSON for programmatic access
• PDF for presentations
• Real-time API access

API Access

Access analytics data programmatically:

// Get usage data
const usage = await unicraft.analytics.getUsage({
  start_date: "2024-01-01",
  end_date: "2024-01-31",
});

// Get cost data
const costs = await unicraft.analytics.getCosts({
  group_by: "provider",
  include_breakdown: true,
});

// Get performance data
const performance = await unicraft.analytics.getPerformance({
  metrics: ["response_time", "success_rate"],
  time_range: "7d",
});

Integration

Webhook Notifications

Receive real-time notifications via webhooks:

const webhook = await unicraft.webhooks.create({
  url: "https://your-app.com/webhooks/unicraft",
  events: ["request.completed", "error.occurred", "budget.exceeded"],
});

Third-party Integrations

Integrate with popular monitoring tools:

• Grafana dashboards
• DataDog metrics
• New Relic monitoring
• Custom monitoring solutions

Best Practices

1. Regular Monitoring: Check your analytics dashboard regularly
2. Set Up Alerts: Configure alerts for important metrics
3. Analyze Trends: Look for patterns in your usage data
4. Optimize Based on Data: Use analytics insights to optimize your AI usage
5. Budget Planning: Use historical data to plan future budgets