Teams | Collaboration | Customer Service | Project Management

AI can sound confident about almost anything. The challenge is making sure it’s right. At Atlassian, Content Technologists are building the systems, standards, and structures that help AI deliver better outputs, removing the burden from our customers of knowing what information to trust.

Earlier this year, we wrote about turning Atlassian Design System guidance into structured content that AI agents can actually use. Since then, agents have reached that content through our MCP server tools and an agent skill. We recently added a third distribution method: a command-line tool. The first question from our own engineers was a fair one: Why ship a CLI when you already have an MCP server?

Atlassian and Gamma are now deeply connected, so your work, and the context behind it, moves with you from plan to presentation. AI is supposed to understand your work. But it can only see what it can reach, and right now, your work is scattered. Updates live in slide decks, details hide in Jira tickets, decisions are buried in Confluence. We expect AI to “just know” how it all connects, then wonder why the output feels generic.

Here’s a stat that’s been rattling around in my head. In our 2026 AI SDLC study, 94% of engineering leaders told us they’re using AI, but only 6% have the systems to actually scale it across their whole software lifecycle. Almost everyone is playing with agents, yet almost no one can let agents run at scale without things breaking.

Tamar Yehoshua is Chief Product and AI Officer at Atlassian, where AI agents are now direct users of the software her team builds.

AI is accelerating implementation. The engineering teams who benefit most will connect context, orchestration, and accountability across the work around code. AI can now help engineers turn ideas into working code in minutes, but software delivery has always been more than just producing code.

AI is changing what entire organizations can accomplish. As it moves beyond individual productivity to orchestrating entire workflows, the value of software outgrows what seat count alone can capture. Atlassian customers are already realizing tangible business value from AI and automations, from reporting processes that run up to 40x faster with agents to saving 400 hours per month with automations.

AI makes it quick and easy to turn rough notes into a clean, easy-to-read document. Job well done, right? Think again. Atlassian’s Teamwork Lab suspected that this professional veneer—what we call “AI polish”—makes it harder to spot foundational flaws and give feedback, so we tested it. We wanted to know: Does AI polish make it harder to identify problems in early drafts? How can we solve for this blind spot?

For many knowledge workers, reaching for AI tools is now automatic. But the Teamwork Lab wanted to know if using AI actually changes how work feels and how much gets done. So Teamwork Lab tested: What happens when you take away AI for a day?

There’s a concept in software engineering called a software factory – a structured, repeatable pipeline that takes raw inputs (requirements, code, tests) and reliably produces high-quality outputs (working software). The magic isn’t just automation. It’s the combination of standard tools, curated configuration, and encoded expertise that makes every run predictable, consistent, and improvable over time.