AI Lens

Simulation with AI Lens

Aisera AI Lens is a tool that allows Administrators to simulate their conversational bot with a webchat channel. NOTE: If you are a user with a custom role, make sure your role has at least the following permissions, which are required to view conversations with AI Lens:

• intent – Read

• channels – Read

• AiseraGPT – Read

Getting Started

• Webchat channel is the default channel associated with the bot

• Select correct bot

• Login with a user

Conversation Window

AI Lens lets you simulate the interaction with your bot. The simulated input Requests are not captured within the Request Details page. AI Lens is similar to runtime analysis.

To use AI Lens:

1. Choose AI Lens in your Aisera Admin UI.

2. Select your bot from the pull-down list at the top of the AI Lens window. You will see a bot interface on the AI Lens window.

3. Add a Request. (AI Lens requests will not be recorded in Aisera analytics.)

1. After you add a Request, you will see the bot functions on the right side of your window.

Bot Functions Show Up on the Right Side

Debug Workflows and Hyperflows Using AI Lens

You can use AI Lens to see the debug information for your bot’s workflows and hyperflows while the bot is running.

To see the debug information for your bot workflow or hyperflow:

1. Perform Steps 1-4 in the section above.

2. Choose the Show More link to see the flowDebugInfo and flowDebugInfo2 methods as they are running.

The flowDebugInfo and flowDebugInfo2 methods are the logs created while your workflow or hyperflow runs.

You can also test Workflows and Hyperflows using the Test the Node button on the AI Automation > AI Workflow Studio window.

Fulfillment Source Per Application/Bot Version

When you run the AI Lens simulation, you are seeing the logs from the conversation server. The AiseraGen AI platform has two different conversation services - the first one (sometimes called the Conversation Server) was for the 1.0 version applications/bot, and the second (sometimes called the ConvAI Server) is an LLM conversation service (that automatically runs a RAG index is for 2.0 application/bots).

The Conversation Server for 1.0 Applications/Bots

If you are simulating/testing a 1.0 Version application/bot, the Fulfillment Source field in AI Lens will contain one of the options that you chose as a source when you created the application/bot.

When the conversation 1.0 application/bot is running, the conversation server tries a couple of the fulfillment engines to answer the user Request, and the AI Lens will display the engine where the answer was found.

The Conversation Server for 2.0 Applications/Bots

If you are simulating/testing a 2.0 Version application/bot, the Fulfillment Source field in AI Lens will display ReasoningEngine.

When 2.0 applications/bots are running, the conversation server will only use the Reasoning Engine from the rag service to do the routing.

AI Lens with Copilot Applications

Applications/bots that are using the Copilot channel default to Copilot mode in AI Lens. Applications without a Copilot channel correctly default to Webchat mode.