Atlassian launches Rovo, its new AI teammate
At its Team ’24 conference in Las Vegas, Atlassian unveiled Rovo, its latest AI assistant designed to enhance productivity. Rovo simplifies data accessibility by aggregating information from both first- and third-party tools, offering a user-friendly AI-powered search tool and seamless integrations into Atlassian’s suite of products. However, the most intriguing feature of Rovo lies in its Rovo Agents, which enable workflow automation in tools like Jira and Confluence. Notably, these agents can be developed using a natural language interface, eliminating the need for programming skills.
Sherif Mansour, Atlassian’s head of product for Atlassian Intelligence, described Rovo as a comprehensive knowledge model for organizations, emphasizing its role as a knowledge discovery tool for all knowledge workers. He highlighted the common workflow loop of finding, understanding, and taking action on work tasks, noting Rovo’s ability to accelerate this process through generative AI technology. This marks a significant milestone in leveraging AI to streamline team collaboration and productivity.
Atlassian Team `24 Las Vegas. Image Credits: Atlassian
Rovo’s foundation lies in Atlassian’s “cloud teamwork graph,” the cornerstone of Atlassian Intelligence, the company’s initiative to integrate AI capabilities into its products, initiated a year ago. This graph aggregates data from Atlassian’s suite of products as well as various third-party SaaS tools.
The proliferation of SaaS tools underscores the need for solutions like Rovo, as each tool typically operates within its own data silo, complicating the process of accessing relevant information for employees. Rovo addresses this challenge by leveraging the cloud teamwork graph to seamlessly consolidate data from multiple sources, thereby enhancing accessibility and productivity for users.
Image Credits: Atlassian
Rovo, as Mansour explained, centers around three key pillars of teamwork: facilitating teams in discovering and engaging with their work, aiding teams in learning, and empowering them to take action effectively.
Enterprise search emerges as the immediate focus, given Atlassian’s existing aggregation of data. This feature promises immediate utility for users by eliminating the need to switch contexts constantly in search of information. Notably, Rovo supports a range of third-party tools right out of the box, including Google Drive, Microsoft SharePoint, Microsoft Teams, GitHub, Slack, and Figma.
Recognizing the diverse tool landscape in enterprises, Rovo offers flexibility for organizations to create their own connectors, tailored to their unique toolsets. For instance, Atlassian developed a connector to integrate its internal developer documentation into Rovo, resulting in significant time savings for developers compared to using AI code-generation tools.
Mansour highlighted that one of the major technical challenges lies in building and maintaining these connectors while ensuring compliance with access permissions set by a company’s IT and security teams. Rovo’s tailored search results respect individual user permissions, ensuring that users only see what they have access to, thereby prioritizing data security and privacy.
Image Credits: Atlassian
In keeping with the trends of 2024, Rovo also debuts as a chat service, leveraging its access to vast data repositories to facilitate retrieval-augmented generation (RAG). This approach involves feeding a large language model with relevant data to provide tailored responses, enhancing user interactions.
While RAG reduces the likelihood of models deviating from the intended script, large language models can still encounter instances of hallucination. To uphold user trust, Rovo meticulously cites its sources, ensuring transparency and accountability. Additionally, interactive previews are often available, particularly with content like slideshows and Figma designs, further enhancing user confidence in the results.
A notable feature integrated into Rovo is its capability to detect and explain company-specific jargon. For seamless accessibility, a Chrome extension is available, automatically underlining and elucidating company-specific terms as users read documents, such as Google Docs. This functionality is driven by Rovo’s semantic search engine, enhancing comprehension and collaboration within organizations.
Virtual teammates
Rovo Agents represent a significant leap beyond simply finding information—they empower users to take decisive action. This extension of Atlassian Intelligence reflects the evolution of virtual teammates, as described by the company.
According to Mansour, Rovo Agents are poised to revolutionize teamwork by leveraging their capacity to process extensive enterprise data, dissect intricate tasks, continuously learn, and collaborate with human counterparts to navigate critical decisions. These agents are not merely advanced chatbots; they possess specialized knowledge and skills tailored to diverse workflows and processes.
Image Credits: Atlassian
Rovo Agents possess a versatile skill set, enabling them to generate, review, and edit content for various purposes, ranging from marketing materials to product specifications and Jira issues. Users can customize agents to address specific inquiries or offer recommendations on best practices. Crucially, these agents can automate tasks triggered by events such as the progression of a Jira issue, assist in Jira backlog management, or facilitate organization of Confluence pages— all while maintaining human oversight.
Mansour expressed a strong conviction in the future of teamwork, envisioning a landscape where human colleagues collaborate seamlessly with virtual teammates—agents. With their integration into daily workflows, users can expect to interact with these agents across a myriad of tasks, enhancing efficiency and productivity in teamwork.