---
title: "Datasets for evaluating an AI feature"
description: "How can you assemble representative tasks for AI-feature evaluation?"
status: "coming-soon"
language: "en"
canonical: "https://bulnik.dev/en/articles/topics/T58.01/"
topicId: "T58.01"
---

> Coming soon — this article is being prepared. Below are its question, intended outcome, and place in the story.

## The question

How can you assemble representative tasks for AI-feature evaluation?

## What we will work through

Assemble representative tasks for AI-feature evaluation.

## Before you begin

Previously covered: A model call, a workflow, and an agent.

- [A model call, a workflow, and an agent](/en/articles/topics/T57.01/)