Imagine troubleshooting your IBM Cloud Pak for Integration (CP4I) environment not with endless log dives or by trying to find your Kubernetes expert for a deep dive, but with a simple natural language question: "What's causing my node to fail?". In a few seconds, the system itself responds, analysing topology, pulling relevant logs, checking versions, and more. You’ll be given a thoughtful response and hear recommended actions to fine tune your environment or fix the issue you asked about. All from within the Cloud Pak for Integration Platform UI. Sounds too good to be true? It's not. Today, I'm thrilled to announce the tech preview of Agentic AI powered Agents for CP4I 16.1.0, a groundbreaking preview of a potential new capability that brings autonomous, AI-powered assistance to your integration operations.

As I've written before in my post on CP4I as the backbone of AI-driven business, integration is all about flowing data from creation to value. But keeping those flows running? Optimising your environment? Just trying to figure out what’s there and what’s not there: that's where admins often hit roadblocks. Manual checks, lengthy resolutions, and the need for all-too-rare expertise becomes a daily headache. The new IBM Cloud Pak for Integration Agent (tech preview) changes the game by leveraging agentic systems: self-directed AI that autonomously solves problems, manages long-term goals, and adapts to your environment. Delivered as an easy-to-deploy new feature that runs directly within your system, interfaced through the Platform UI. It works for cloud-hosted or on-premises runtimes, with all access secured through CP4I and OpenShift RBAC.

Let's dive into the more information about what these Agents can do for you. As the user, you interact with the Agents through the chat window in the Platform UI. Now, at your fingertips is an army of agents, each specialised in providing information that will be helpful. The agents will work to answer your queries with the following skills:

1.         Take your query and farms out the question, and pulls together the responses, providing clear information.

2.         Parsing the information in the product documentation, telling you just what your need to know.

3.         Mapping your deployment topology and understanding relationships between resources to understand the big picture.

4.         Skilled information and understanding about Kubernetes to help resolve potential issues.

5.         Reviewing information in the most relevant logs, highlighting what's interesting.

6.         Checking what versions are currently installed and providing recommendations to assist either optimal deployments or to address critical issues.

This isn't just automation, or even just timesaving. It’s intelligent, proactive reliability. Picture health checks spotting anomalies before they escalate, or upgrade suggestions that prevent downtime. And the best part? It's all transparent: You see exactly what the agents are analysing, why, and how, with full traceability for accountability. And you don’t need to spend your time diving into menus, or logs or system files to try and anticipate what might be happening. You can simply ask.

The potential benefits are game-changing. Admins gain instant expertise, slashing resolution times and boosting productivity. Imagine the benefits in resolving issues in minutes instead of hours. Think of all the additional useful work that can be done in the hours that have been freed up! Your organization stays ahead with detection anomalies and takes advantage of recommendations for improved deployments, enhancing overall satisfaction. Early customers who have already been briefed about this have responded with excitement and a desire to try this quickly. For businesses taking advantage of CP4I's container-native architecture (as I explored in my blog on “Precision in integration”), these agents boost that agility, making operations smarter and more efficient.

And here's the thrilling news: This tech preview is available at no cost for CP4I 16.1.0 users. Plus, if you sign up for this early access, you will get watsonx.ai tokens to power these agents. This is your opportunity to experience the future of integration ops firsthand and provide feedback to shape what's next.

Ready to transform your CP4I experience?

Visit the sign-up page at http://ibm.biz/cp4i-agent-preview

This will outline how you get access to the Agents, how to use them, and how to provide your valuable feedback.

For questions and to get access to a dedicated Slack channel email us on AskIntegration@ibm.com

And for even more insight, join us for a webinar covering not just the Cloud Pak for Integration Agents but also new Agents for IBM MQ and IBM App Connect as well. Sign up here.

Whether you're dealing with complex integrations or just curious about AI's role in ops, this preview is a must-try. What's the most compelling operational challenge you'd love AI to solve? We would love to hear from you and to see whether these agents are just what you need. Just try and stay focused!