Tutorials

Reproducible example use cases for DBNL

This section contains examples demonstrating how to use DBNL in various scenarios. Each example includes working code, detailed explanations, and guidance on when to use each approach.

All of these tutorials can be previewed in our Read Only SaaS environment.

ADK Calculator Tutorial

The ADK Calculator Tutorial provides a comprehensive walkthrough of building an end-to-end analytics pipeline:

• Generate OTEL traces from a Google ADK calculator agent

• Convert and augment trace data with computed metrics

• Upload multi-day trace data to DBNL

• Analyze agent behavior over time

A/B Testing Tutorial

The A/B Testing Tutorial demonstrates how to compare agent versions:

• Upload traces from multiple agent versions with cohort labels

• Add comparison metrics like accuracy and error rates

• Use DBNL segmentation to analyze version differences

• Validate improvements before full rollout

Hyperparameter Optimization Tutorial

The HPO Tutorial shows how to analyze hyperparameter impact:

• Upload traces from agents with different configurations

• Compare pre and post-optimization performance

• Use DBNL insights to validate optimization results

• Build automated optimization pipelines with DBNL feedback

Repository

All example code is available in the dbnlAI/examples GitHub repository.