Skip to main content

Analytics Dashboard

Get deep insights into your AI systems with comprehensive analytics covering usage, costs, compliance trends, bias detection, and performance metrics.

Key Metrics

Usage

API calls, tokens, and request patterns

Costs

AI provider costs and budget tracking

Compliance

Scores, violations, and trends over time

Performance

Response times, success rates, errors

Bias Detection

Fairness metrics and bias indicators

Risk Assessment

Risk scores and Governor validations

Usage Analytics

API Gateway Metrics

Track all AI requests through RegPilot Gateway:
  • Total Requests: Requests per day/week/month
  • Tokens Used: Input and output tokens
  • Model Distribution: Usage by AI model (GPT-4, Claude, etc.)
  • Response Times: Average, P50, P95, P99 latency
  • Success Rates: Successful vs failed requests
  • Error Types: Breakdown of error causes

Cost Tracking

Monitor AI spending:
{
  "period": "2024-01",
  "total_cost": 1247.32,
  "breakdown": {
    "gpt-4": 892.15,
    "gpt-3.5-turbo": 245.87,
    "claude-3-opus": 109.30
  },
  "budget": 2000.00,
  "usage_percent": 62.4
}

Compliance Analytics

View compliance score changes over time:
  • Daily/weekly/monthly compliance scores
  • Violations by type and severity
  • Framework-specific compliance (EU AI Act, GDPR)
  • Resolution rates and times

Bias Detection

Advanced fairness metrics:
Measures whether predictions are independent of protected attributes
Checks if true positive rates are equal across groups
Ensures predicted probabilities match actual outcomes
Similar individuals receive similar predictions

Performance Analytics

Response Time Distribution

Monitor AI response performance:
  • P50 (median): Typical response time
  • P95: 95% of requests faster than this
  • P99: 99% of requests faster than this
  • Max response time
  • Timeout rate

Model Performance

Track model accuracy and reliability:
  • Success rate by model
  • Error rates and types
  • Retry patterns
  • Governor validation rates

Custom Reports

Create custom analytics reports:
1

Select Metrics

Choose which data points to include
2

Set Filters

Filter by project, model, date range, etc.
3

Choose Visualization

Select chart types (line, bar, pie, table)
4

Schedule & Export

Email reports or export as PDF/CSV

API Access

Access analytics programmatically:
// Get usage analytics
const usage = await regpilot.analytics.usage({
  startDate: '2024-01-01',
  endDate: '2024-01-31',
  groupBy: 'day'
});

// Get compliance trends
const compliance = await regpilot.analytics.compliance({
  framework: 'eu_ai_act',
  period: 'last_30_days'
});

// Get cost breakdown
const costs = await regpilot.analytics.costs({
  currency: 'USD',
  groupBy: 'model'
});

Alerts & Notifications

Set up analytics-based alerts:
  • Budget threshold exceeded
  • Unusual usage patterns
  • Performance degradation
  • Compliance score drops
  • High error rates

Next Steps