Atlassian today added a series of artificial intelligence (AI) agents that are specifically trained to automate a range of software engineering tasks.
Announced at the Atlassian Team’25 conference, these additions to the Atlassian portfolio of Rovo agents, now available in beta, include a code planner that turns Jira tickets into technical plans, agents that write code, an agent that reviews code and a deployment agent.
Atlassian is also adding a context-aware search engine that enables software development teams to launch natural language queries to, for example, discover source code with Jira issues, architecture documentation and technical discussions.
Finally, Atlassian is also making available Rovo Studio App, a tool for building agents, automations and other types of applications using a mix of natural language and traditional coding techniques.
Chirag Shah, head of product for developer solutions for Atlassian, said that approach makes it simpler for application developers to mix and match tools as they best see fit. AI is not so much a game changer that eliminates the need for application developers so much as they are a set of tools and capabilities that automate manual tasks, he added.
The overall goal is to make it simpler for developers to write code in a way they still easily understand and, by extension, trust, noted Shah. That can be critical because many developers are struggling to debug code created by AI coding tools simply because they lack any insights into how it was created.
Mitch Ashley, vice president and DevOps and application development practice lead for The Futurum Group, said Rovo Studio App is a workbench where AI agents leverage contextualized data necessary to perform work across the software development lifecycle. In effect, we’re now seeing AI agents leveraging knowledge graph technologies to provide higher levels of context, he added.
Similar to other Rovo agents, the application development agents that Atlassian is providing make use of a knowledge graph framework and AI models that the company is integrating across its entire application portfolio. Eventually, Atlassian plans to make available a broad portfolio of AI to automate a wide range of tasks across the software development lifecycle, said Shah.
It’s not clear how widely AI is being employed across the software development lifecycle, but a Futurum Research survey finds that 41% of respondents expect generative AI tools and platforms will be used to generate, review and test code, while 39% plan to make use of AI models based on machine learning algorithms. More than a third (35%) also plan to apply AI and other forms of automation to IT operations, the survey finds.
There is little doubt that AI tools are about to transform the way applications are built and deployed. The issue now is determining how best to incorporate AI agents into software engineering workflows in a way that enables humans to focus more on creative tasks versus all the manual ones that today conspire to often make building applications unnecessarily tedious.
