Back to Cloud and Data Viz

Watson AIOps

What is Watson AIOps?

This product integrates with the customer’s existing IT infrastructure to provide a holistic view of their environment.
Using AI, insights are delivered via the customers’ existing collaboration tools like Slack or MS Teams. The goal, however, is to train the product to automate incident remediation so the user can focus on more important tasks.

How does it do this? Watson AIOps integrates with tools like SericeNow, Dynatrace, LogDNA and more. It monitors incoming logs, metrics, alerts and tickets to highlight existing and potential problems. It uses machine learning to make recommendations and empowers the customer with the ability to see what’s ahead and avoid incidents and outages before they happen. Learn more.

My role and business objective

In order for Watson AIOps to provide such insights, it must first integrate with the user’s estate. My first role on this project was to create a user experience for that integration. I new we had customers with thousands of Github repos that would need to be integrated and scanned, so my work was cut out for me.

After injection of the customer’s infrastructure and tooling, we had to provide a visualization of the customer’s estate and show them any immediate vulnerabilities they may be facing. Working on a team of other UX designers and Researchers, we integrated with the dev team to leverage existing IBM technology in our portfolio. By taking the technological underpinning in existing products like Netcool, we created such a visualization.

How I practiced IBM Design Thinking

Working with Research we had to gather a firm understanding of the personas involved with the entire lifecycle. We also included Watson as a persona since it plays a critical role in the process.

Looking at all of the data involved in software development and incident remediation, we noticed that to understand what constitutes an incident, Solution Architects must write “policies.” This was a pain-point for the SAs. Using machine learning, once Watson learned the customer’s enterprise, it could make recommendations for policies. The SA can customize the policies to be tailored to the needs of the enterprise, therefore applying human override to the AI, which is always essential.