Skip to main content
AI agents don’t just answer questions—they engage in complex, multi-turn conversations to achieve goals. Netra’s Simulation framework lets you test these interactions systematically, simulating realistic users to validate your agent’s performance before deployment.

Quick Start: Simulation

New to simulations? Get your first simulation running in minutes.

​
Why Simulation Matters

Traditional testing falls short for conversational agents. Simulations provide a comprehensive way to test multi-turn interactions with realistic user behaviors:

​
Text and Voice Simulations

Netra supports simulation of multiple types of AI agents, including text agents and voice agents. Simulations come in two modalities. You choose the modality when creating a multi-turn simulation:

Text Simulations

A simulated user chats with your agent over text. You define the scenario goal, behavior instructions, a persona, and facts the agent must communicate.

Voice Simulations

A persona calls your agent over a real audio channel. Configure the STT, reasoning LLM, and TTS pipeline, add background noise, and evaluate how the call actually sounded.
Both modalities share the same building blocks—evaluators, simulations, and test runs—and produce the same transcripts-with-scores workflow.

​
Core Building Blocks

The Simulation suite is built on three interconnected pillars:

​
Evaluators

Evaluators assess the conversation after it completes. Library evaluators are organized by category: Session-level Quality and Agentic evaluators apply to both modalities with a default pass threshold of >= 0.6. The TTS, STT, and Conversational categories score voice-specific qualities such as voice naturalness and transcription accuracy.

​
Simulations

Simulations are collections of simulation scenarios that define multi-turn conversation goals.

​
Test Runs

Test Runs execute your simulation scenarios, providing detailed transcripts and evaluation results.

​
Getting Started

1

Configure Evaluators

Set up evaluators to define your scoring criteria — choose from the library or create custom ones.
2

Create a Simulation

Build a multi-turn simulation with simulation scenarios, personas, and context data—for text or voice.
3

Run Simulations

Execute your simulation from the dashboard or SDK and review transcripts and results in Test Runs.
4

Analyze and Improve

Use insights from simulations to refine your agent’s behavior.
Last modified on August 28, 2026