Relevance AI: agent building features and setup

Overview

Relevance AI presents agents that specialize in tasks such as company research, meeting preparation, call summaries, and business information updates. Instead of one general agent for the entire organization, you can define a focused role and integrate it into a team workflow. This approach suits recurring tasks with results that can be checked. To decide whether to expand usage, measure quality, cost per task, and how much correction a person needs to make.

Features

• Specialized agents for sales, marketing, service, and operations roles.
• Tailoring agents to a team's tasks and tools.
• Tracking task performance, quality evaluations, and usage.

Business uses

• Preparing a summary for a customer meeting using authorized information sources.
• Researching a company and completing its data before a salesperson checks its fit.

Getting started

• Choose a task with a clear output and write an example of a good result.
• Set the sources and tools, then evaluate a sample of tasks in a test environment.
• Compare the results with the team's work and decide when human review is required.

What to check

• For company research, keep sources and check dates so outdated information does not make its way into the CRM.
• Before expanding an agent to writing or sending tasks, verify permissions and approval for every external action.