Atlassian unveiled Rovo, a groundbreaking AI assistant, during its Team ’24 conference in Las Vegas. Designed to streamline information access, Rovo utilizes data from both first- and third-party tools through a new AI-powered search feature and other integrations across Atlassian products. A standout feature is the Rovo Agents, which enable users to automate workflows in tools like Jira and Confluence with ease. Impressively, anyone can create these agents using a natural language interface—no coding skills required.
“We envision Rovo as a comprehensive knowledge model tailored for organizations. It serves as a knowledge discovery solution for every knowledge worker,” stated Sherif Mansour, Atlassian's head of product for Atlassian Intelligence. “Every desk worker goes through a process of finding, understanding, and taking action on tasks. Rovo represents a significant leap in generative AI, enhancing our capabilities for team efficiency.”
Rovo is built on Atlassian’s “cloud teamwork graph,” which underpins Atlassian Intelligence—a year-old initiative aimed at integrating AI into its suite of products. This graph consolidates data from Atlassian's tools and various third-party SaaS applications, addressing the challenges posed by data silos often found in multiple tools. This need highlights Rovo's value, simplifying access to information across the enterprise.
Mansour emphasized that Rovo is built around three core teamwork principles: assisting teams in finding and connecting with their work, facilitating learning, and enabling action.
Enterprise search represents an immediate win for users, allowing them to avoid the hassle of switching between tools. Rovo comes with out-of-the-box support for popular third-party applications such as Google Drive, Microsoft SharePoint, Microsoft Teams, GitHub, Slack, and Figma. Enterprises can also create custom connectors; for instance, Atlassian developed a connector for its internal developer documentation, estimated to save developers an hour or two each week.
Mansour noted that the primary technical challenge involves creating these connectors and ensuring they adhere to company IT and security protocols. Rovo ensures personalized search results based on user permissions, allowing individuals to see only what they are authorized to access.
Furthermore, Rovo integrates a chat service, leveraging retrieval-augmented generation (RAG) to deliver tailored responses powered by a large language model. This approach minimizes the likelihood of inaccuracies, and Rovo transparently cites its sources, often including interactive previews for better context.
An intriguing feature of Rovo is its capability to identify and explain company jargon. A browser extension highlights and clarifies field-specific terms while users read documents, leveraging Rovo’s semantic search capabilities.
Rovo Agents enhance the platform by facilitating action based on the information found. Described as “virtual teammates,” these agents synthesize vast amounts of enterprise data, simplify complex tasks, adapt through continuous learning, and collaborate with human team members for critical decision-making.
“Rovo Agents are not merely advanced chatbots; they possess specialized knowledge applicable across various workflows,” Mansour explained. These agents can generate, review, and edit content for marketing, product specifications, or Jira issues. They can also automate tasks tied to Jira workflow stages, assist with decluttering Jira backlogs, and organize Confluence pages—all while ensuring human oversight.
Mansour reinforces a compelling vision: “We believe the future of teamwork lies in collaboration between human employees and their virtual counterparts—agents.”
Ultimately, Rovo stands to revolutionize how teams manage information and work together, enhancing productivity through smarter workflows.
Ranking
