AI in Software Development
DORA Report Reveals Seven Team Profiles and Seven Capabilities for Maximizing AI Value in Software Development
New research identifies how organizational systems, team structure, and specific practices determine whether AI amplifies productivity or creates isolated pockets of dysfunction.
InfoQ
The latest research from DORA highlights a critical insight: AI acts as an amplifier of existing organizational strengths and weaknesses, rather than a standalone solution. According to Nathen Harvey's presentation at QCon London, the greatest returns on AI investments come not from the technology itself, but from intentional focus on the systems, processes, and culture that underpin software development teams.
Seven Distinct Team Profiles#
DORA's 2025 research identified seven unique team profiles that respond differently to AI adoption:
- Foundational challenges
- The legacy bottleneck
- Constrained by process
- High impact, low cadence
- Stable and methodical
- Pragmatic performers
- Harmonious high-achievers
Among these, the "constrained by process" profile exemplifies the challenge: while teams show strong delivery stability, they experience high burnout and friction, low individual effectiveness, and limited time spent on valuable work. Without intervention, AI adoption in such teams risks deepening existing problems rather than solving them.
Seven Critical Capabilities Drive Better Outcomes#
DORA identified specific practices that, when paired with AI adoption, unlock genuine value creation:
- Clear and communicated AI stance
- Healthy data ecosystems
- AI-accessible internal data
- Strong version control practices
- Working in small batches
- User-centric focus
- Quality internal platforms
The Paradox: Higher Productivity, Lower Stability#
Interview data reveals a striking tension: as organizations adopt AI, they report increased individual effectiveness and improved code quality. Yet simultaneously, software delivery instability rises—deployments are rolled back more frequently, and hot fixes become necessary. Harvey attributes this to misaligned incentives where engineers focus on feature velocity without sufficient visibility into production outcomes.
