SmartBear at Atlassian Team ’26: AI, quality intelligence, and the System of Work 

For everyone in QA inside the Atlassian ecosystem, the last few months have made one shift obvious. AI is changing how software gets built, and the quality function must change with it. 

That was the throughline of Atlassian Team ’26 in Anaheim earlier this month. Atlassian is restructuring its platform and ecosystem strategy around AI-enabled workflows powered by shared operational context, with the Teamwork Graph, Rovo, MCP, and agent-based workflows now positioned as the foundation of its System of Work strategy. The message across keynotes, roadmap sessions, and partner discussions was consistent: AI without context has limited value, while AI connected to workflows, organizational data, and automation can accelerate delivery and decision-making across the enterprise. 

That direction is one we have been building toward at SmartBear for some time. Team ’26 was the moment we saw it reflected in almost every conversation we had on the event floor. 

The customer conversation has changed 

A few years ago, most booth conversations at an Atlassian event would have started with a question about test cases or coverage. This year was different. 

The hundreds of people who came to our booth weren’t all traditional QA practitioners. Engineering managers wanted to talk about AI-generated requirements. Delivery leads asked about no-code automation. Platform owners came with questions about agent-driven release readiness. And several customers, framing it in different ways, asked one underlying question: how do we make quality part of delivery? 

This is the question SmartBear Zephyr Agent for Rovo and the SmartBear MCP Server are built to answer, and they answer it for different people. Zephyr Agent for Rovo brings AI-powered quality intelligence into Jira for QA practitioners and managers, surfaced through conversation. It connects to the work items teams already use and runs inside the surface they already work in. 

The SmartBear MCP Server takes Zephyr’s test data outside of Jira, into developer tools and agentic pipelines like Claude, Cursor, and ChatGPT. Developers can pull quality intelligence directly into the environment they’re already coding in, without leaving it. 

The result is that quality intelligence shows up wherever the work happens, for whoever is doing it. 

Testing as part of delivery to move at AI speed and scale 

A second pattern across the week was that teams are scoping testing differently. It’s being treated as part of how teams accelerate delivery, embedded in the workflow rather than executed at the end of it. 

This is the work we describe as application integrity: continuous assurance that your software just works as intended, with governance built in for the speed of AI-driven development. Releases moving at AI speed need testing that moves at AI speed, with traceability, automation, and team-wide visibility built into the workflow. 

Zephyr Skills for Rovo brings coverage analysis, release risk assessment, and test case generation into Jira directly, surfaced through conversation. The SmartBear MCP Server extends this further by letting custom Rovo agents trigger no-code automation, update Jira work items, and notify teams in a single conversation. 

This is how QA teams keep pace with delivery instead of trailing behind it. 

Data Center-to-cloud migration was a common conversation topic 

Outside of AI, Data center-to-cloud migration was top of mind for many of the customers we spoke with. Atlassian’s roadmap signals make it clear that the next wave of innovation around Rovo and the Teamwork Graph will land on Atlassian Cloud, and customers are planning accordingly. 

SmartBear is aligned with this direction. Our work with Atlassian customers is increasingly focused on making the Cloud transition feel like a managed partnership, with the operational coordination and clarity enterprise teams expect. If you are a Zephyr customer planning or already in the process of moving to Cloud, our support team is ready to work through it with you.  

Jira admins are central to the testing conversation 

Jira admins and platform owners are increasingly central to how testing tools get evaluated and deployed inside Atlassian accounts. Knowing that they are gatekeepers responsible for recommending what goes on their systems, many admins visited us at Team ’26 to learn more about how Zephyr could help the testing teams they service.  

When quality is embedded in the workflow and governed at the platform level, the people who own the platform have a direct stake in which tools make it in. Their questions about implementation, security, scalability, and operational fit are the right questions, and the answers matter as much to a successful rollout as the feature set does. 

Zephyr is built with this in mind. The architecture, deployment model, and documentation are designed to give platform owners the confidence they need to move forward. 

Atlassian, Solution Partners, and the push to put AI to work 

Team ’26 made clear that getting AI to work for every team involved in software delivery is a shared priority across the ecosystem. Atlassian and its Solution Partners are stepping up to help customers understand how to put Rovo and AI-powered workflows to work in their delivery pipelines, and we’re looking forward to working with them to make that happen. 

SmartBear is building with this ecosystem; that’s what this partnership demonstrates. That work was recognized mid-week when Atlassian named SmartBear Partner of the Year 2026: AI Innovator, in recognition of the Rovo-powered work within Zephyr that brings conversational AI-driven quality intelligence directly into Jira. 

If you want the full story behind the recognition, we wrote about it here. 

What this means for QA teams inside Jira 

The conversations at Team ’26 reinforced what teams already know. The release readiness question is getting harder to answer with manual coordination, and AI changes both the pace of delivery and what is possible inside the quality function. 

For QA teams using Zephyr today, the path forward is already in your hands: 

• Faster time to insight on release readiness 

• Improved test coverage and reduced risk exposure 

• Streamlined workflows with less context switching 

• Stronger confidence in delivery decisions 

Get started with Zephyr on the Atlassian Marketplace 

The Rovo-powered capabilities discussed throughout this post are live and ready to use. Explore the Zephyr listing on the Atlassian Marketplace to see how Zephyr serves as your testing system of record with AI-driven quality intelligence built in. 

If you want to see Zephyr Skills for Rovo in action, this short demo walks through how it brings test management and automation insights directly into Jira.