DOCSEARCHbyAlgolia
Submit Docs

Agentic CX Designer

website·Indexed 35 days ago

https://developers.nlx.ai

A service within Amazon Connect Customer that works with Connect Customer flows to design and improve customer experiences using real conversation performance.

Tokens
645
Snippets
0
Records
4
Agent score
3%

What's inside Agentic CX Designer

  1. Overview of Agentic CX designer

    Agentic CX designer is a service within Amazon Connect Customer used to create conversational AI applications. Unlike simple bots, these applications can guide structured workflows, answer questions from trusted knowledge bases, connect to APIs, reason through multi-step tasks, and escalate to human agents. It provides a single workspace for the entire conversational AI lifecycle: building, testing, deploying, observing, and optimizing.

  2. Integrate Agentic CX designer with Connect Customer flows

    Agentic CX designer acts as the conversational AI application layer, while Connect Customer flows manage the customer entry point and routing. In a typical implementation, a Connect Customer flow uses an 'Agentic CX block' to invoke a specific conversational AI application and its deployed environment.

    Division of Responsibilities:

    • Connect Customer flows: Handles contact flows, routing, queues, channel setup, and the final escalation into the contact center experience.
    • Agentic CX designer: Handles application logic, nodes, prompts, integrations, knowledge bases, guardrails, and determines when to trigger an escalation.

  3. Utilize built-in AI capabilities in conversational applications

    Agentic CX designer includes several generative AI capabilities to enhance user interaction:

    • Text Generation: For dynamic responses, summaries, and explanations.
    • Intent Classification: Routing users to the correct flow based on AI descriptions of the utterance.
    • Slot Capture: Naturally collecting multiple pieces of information, even if provided out of order.
    • Generative Conditions: Routing based on semantic meaning rather than exact value matches.
    • Data Transformation: Filtering, mapping, or reshaping data before presenting it to the user.
    • Agentic Reasoning: Using agent nodes to complete multi-step tasks via tools and knowledge.
    • MCP-enabled flows: Allowing structured flows to be used as tools by an agent.

  4. Core capabilities of Agentic CX designer

    Agentic CX designer provides several tools for building complex AI experiences:

    • Visual Workflow Design: A no-code Canvas for building structured conversation paths.
    • Deterministic Logic: Predictable steps, decisions, and API calls.
    • Generative AI Nodes: Dynamic response generation, intent classification, and data transformation.
    • Agentic Experiences: Agent nodes that reason through multi-step tasks and call tools.
    • Knowledge Bases: Grounding AI responses in trusted content.
    • Integrations: Connection to managed services and custom APIs.
    • Guardrails: Safety, brand, and compliance controls for inputs and outputs.
    • Testing & Analytics: Event logs, state inspection, and performance dashboards.