Examples

Example Workflows to Showcase Evaluation Power

In this section, we will present a series of examples that illustrate how to create workflows designed to highlight the evaluation capabilities of Dynamiq. These workflows will serve as practical demonstrations of how to leverage evaluation metrics effectively in different scenarios.

Upcoming Examples

The following subpages will feature detailed examples for creating various types of workflows, including:

1. Accurate Workflow:

   • Learn how to set up a workflow that consistently generates accurate answers, allowing you to evaluate the effectiveness of your metrics.

2. Inaccurate Workflow:

   • Explore the creation of a workflow that produces intentionally incorrect answers. This workflow will help demonstrate how well your evaluation metrics can identify inaccuracies.

3. RAG Workflow:

   • Gain insights into building a Retrieval-Augmented Generation (RAG) workflow, showcasing the integration of retrieval and generation processes along with their evaluation.

4. Benchmarking LLMs:

   • Understand how to create workflows for benchmarking different Large Language Models (LLMs), providing a framework for comparing their performance across various tasks.

Each of these examples will provide step-by-step instructions, allowing you to replicate and adapt the workflows to your specific needs. By following these guides, you'll gain a deeper understanding of how to utilize Dynamiq’s evaluation framework to enhance your AI applications.