Atlassian's analysis of AI-native software development highlights a 19% increase in merged pull requests and 2–3 hours saved per developer each week by integrating AI tools throughout the development lifecycle.
- AI tools boost merged pull requests by 19%
- Developers save 2–3 hours weekly on coding tasks
- Full SDLC is evolving with continuous human-agent collaboration
What happened
Atlassian’s Rovo Dev AI platform, which integrates AI capabilities into coding and code review, has demonstrated measurable productivity improvements across thousands of repositories. The data shows developers are merging 19% more pull requests compared to previous workflows without AI assistance. This increase is accompanied by developers saving between two and three hours per week on average by automating routine coding and review tasks.
The analysis reflects a broader trend where 93% of developers use AI tools and nearly 30% of code today is AI-authored. The AI-native Software Development Lifecycle now spans planning, coding, reviewing, testing, and operations, enabling teams to speed up delivery while maintaining or improving quality.
Why it matters
These findings highlight how AI is reshaping software development beyond just helping write code. Developers are reinvesting saved time into improving code quality, documentation, and engineering culture. However, as AI accelerates coding outputs, bottlenecks in planning, reviews, and integration become more apparent, underscoring the need for a holistic approach to AI in SDLC.
The shift towards an AI-native SDLC signals a new collaboration model where AI agents and human developers work in feedback loops at every stage, rather than a traditional linear handoff process. Understanding where AI adds value and measuring its true impact on productivity remain key challenges as organizations scale AI adoption.
What to watch next
Future developments will focus on expanding AI’s role across all SDLC phases including planning, testing, and operational monitoring, with an emphasis on orchestrating workflows that integrate human insight and AI automation seamlessly. Teams will likely require more sophisticated dashboards that track not only AI usage but also the quality and impact of AI-generated contributions to software projects.
Engineering organizations will be monitoring if the productivity gains from AI tools persist over time and translate into higher quality software and improved developer satisfaction. The broader industry will watch how AI’s contribution evolves from being a code generator to a holistic partner in software delivery, driving a new era of developer experience innovation.