Tabnine and Atlassian reveal new generative AI tools for developers

Just days after Github’s Copilot Workspace launch, Tabnine and Atlassian are adding to the growing number of AI-powered support tools available to developers. 

Tabnine is a generative AI code assistant built to support developers and boost productivity, and Atlassian offers a suite of team collaboration tools. On Wednesday, the companies co-launched a preview of coming AI integrations, part of Atlassian Rovo, at the Atlassian Team 2024 conference. 

Also: I tested the AI gadget that got the internet buzzing and it left me wanting more

The integrations will “allow Tabnine’s customers to leverage the complete body of information — the data and code that represents deep institutional knowledge — that’s captured across Atlassian’s suite of products,” including Jira, Confluence, and Bitbucket, according to the release. The updates will be available later this year, though the announcement did not specify a date. 

The companies also shared in the announcement that Tabnine’s AI coding assistant will be embedded across Atlassian’s entire product suite, “to provide the company’s full scope of generative AI features anywhere an engineering team does their work.” 

“Our extended partnership with Atlassian is a natural step forward in supporting engineering teams anywhere and any way they do their work, empowering them with cutting-edge AI capabilities within the Atlassian ecosystem,” Tabnine CEO Dror Weiss said in the release. “Our focus is on providing engineering teams with flexibility and control, ensuring that they can adopt AI within how they develop software without having to change the other tools they use to see the full benefit.”

Also: The best AI chatbots: ChatGPT and alternatives

In a related announcement on Tuesday, Tabnine shared a preview of an “Atlassian Jira-to-code AI agent” that uses natural language to fulfill a Jira ticket and build a functional application with context from an existing project. 

The Rovo integrations build on Tabnine and Atlassian’s longstanding relationship. Tabnine is already integrated with Bitbucket, Atlassian’s code hosting and collaboration platform, which helps Tabnine’s AI assistant personalize its code recommendations. 

“Personalized AI recommendations based on awareness of a developer’s code are accepted 40% more often than AI recommendations generated without these integrations,” Tabnine said in the release. 

Also: You can make big money from AI – but only if people trust your data

The Team 2024 announcement adds to Tabnine’s recent releases, which also support Atlassian users. Tabnine upgraded customization in February with local code awareness; currently in preview for Tabnine Enterprise customers, the feature lets users connect Tabnine to their team’s Bitbucket, Github, or GitLab repository for “more personalized, higher quality results when generating code, explaining code, creating tests, writing documentation, and more.” 

Last month, Tabnine announced switchable models, which allow users to choose the large language model (LLM) that powers Tabnine Chat, the company’s AI developer chatbot. At the time of writing, users can select from Tabnine’s original model, Tabnine + Mistral, GPT-3.5 Turbo, or GPT-4.0 Turbo, and switch between them in real time.