Agentic AI in Self‑Service Portals: Guide users to complete cases seamlessly

User Experience
1

The article walks you through designing and implementing a Coversational Agent rule and includes a demo you can adapt to your own use case.

Conversational Agent:

  • It’s not just a chatbot. It’s an assistant that can plan multi‑step tasks, call business tools (APIs, rules, document extractors), remember context, and take actions on the user’s behalf to move a case forward.
  • The agent recognizes intent, gathers what’s missing, executes steps (create/update case, verify identity, collect documents, schedule, take payment), and confirms outcomes—escalating to a human when needed.

Details about Pega conversational Agent

  • Agents are GenAI-powered assistants embedded in cases that guide users through workflows using natural language.
  • Designers use the new Agent and Tool rule types to define conversation capabilities and the automations/integrations the Agent can invoke.
  • At run time, the Agent interprets free‑form requests, selects the right tools, fetches or updates data, and advances the case automatically.
  • The result is a guided, personalized experience that accelerates resolution, reduces manual effort, and adapts to each user’s intent.

Demo: In this demo, a Pega Conversational Agent guides applicants through a personal loan case, prompting for required details at each step. Users complete the workflow via clear, guided prompts, and the Agent only executes assignments it is authorized to complete.

https://players.brightcove.net/1519050010001/default_default/index.html?videoId=6385977568112

Steps:

  1. In App Studio → AI Designer → Agent and create new.
  2. Provide the Class, Name and description of the Agent.
  3. AutoPilot generates the necessary artifacts based on the description provided for Agent rule.
  4. Define the scope: Application for the Agent rule.
  5. Go to Case Types tab - Add the Case Types
  6. The Conversation agent can be added to Self Service Portal or Landing Pages (screenshot below for your reference).

Conversational Agent.jpg
Conversational Agent.jpg3316×1816 291 KB

Next steps: The next article will show how to configure Tools and Knowledge and how Agents use them in real time. We’ll cover setting up Tool rules, connecting to data and actions, defining Knowledge sources, and best practices for safe, scalable orchestration. Stay tuned for a step‑by‑step guide you can apply immediately.

Reference:

https://docs.pega.com/bundle/platform/page/platform/gen-ai/automating-work-with-agents.html

https://docs.pega.com/bundle/platform/page/platform/gen-ai/creating-agents-app-studio.html

https://docs.pega.com/bundle/platform/page/platform/gen-ai/exposing-pega-genai-run-time.html

Pega GenAI Cookbook - Recipes series

Enjoyed this article? See more similar articles in Pega Cookbook - Recipes :fire::fire::fire: series.

2

@RameshSangili - it will be good if you add the link to the next article in this series in this post itself. Thanks

3

@Atanu Senhi we added link at the bottom of article to main article that lists all gen ai related articles.