Before sending any records you must initialize the logger. Initialization tells Quotient which app generated the data, which environment it came from, and which detections to run.

​
Basic Initialization

from quotientai import QuotientAI, DetectionType

quotient = QuotientAI(api_key="your-quotient-api-key")

quotient.logger.init(
    app_name="my-first-app",
    environment="dev",
    detections=[DetectionType.HALLUCINATION, DetectionType.DOCUMENT_RELEVANCY],
    detection_sample_rate=1.0,
)

​
Key Parameters

  • app_name (string): Logical application name. Used as the primary grouping dimension in the dashboard.
  • environment (string): Where the app is running (dev, staging, prod, etc.). Drives environment filters.
  • tags (object, optional): Default metadata applied to every log (e.g., { "team": "growth" }).
  • sample_rate (float): Percentage of logs to persist. 1.0 captures everything, lower values sample.
  • detections (array): Which automated checks to run, such as HALLUCINATION or DOCUMENT_RELEVANCY.
  • detection_sample_rate (float): Sampling rate for detections specifically. Defaults to 0.0 (disabled).

​
Tips

  • Call logger.init once during application startup; repeated calls reuse the same configuration.
  • Use configuration files or environment variables to keep app name and environment consistent across deploys.
  • Provide tags for metadata you want on every log (e.g., team, product surface).
Next: Send Logs