> ## Content Index > Fetch the complete content index at: https://www.stirup.si/llms.txt > Use this file to discover other available public pages before exploring further. # Uneven AI Agent Adoption Among Developers - URL: https://www.stirup.si/uneven-ai-agent-adoption-among-developers-5963f24e/ - Published: 2026-10-13T14:45:17.000Z - Updated: 2026-10-13T14:45:17.000Z - Description: Engineering organizations report uneven AI agent adoption, with 20% of engineers generating 80% of code, highlighting productivity disparities and challenges in team collaboration. - Author: Satish Das - Tags: Development, News, Stirup Research ## Uneven Adoption of AI Agents Among Developers Over the past few months, engineering organizations—from small software agencies to large telcos and game studios—have been exploring the impact of AI agents on development workflows. While individual developers can now move much faster with these tools, the gains are uneven. An engineering manager at a mapping company highlights this disparity, noting that 20% of the engineers are generating 80% of the code or consuming 80% of the tokens. This uneven adoption suggests that while AI agents can boost productivity for some, they may not be universally adopted or effectively integrated across teams. The result is a growing divide in how different developers leverage these tools, which can affect overall team velocity and collaboration. As organizations continue to evaluate AI agent adoption, addressing this imbalance will be key to ensuring broader team benefits. ## Challenges in Sharing Knowledge and Best Practices Faster developers don’t necessarily translate to a faster team, as challenges in sharing knowledge and best practices often hinder collective progress. A staff engineer at a financial data provider notes, “There’s a big gap in proficiency,” highlighting that differences in skill levels may not always stem from intelligence but from inconsistent practices or lack of training. Capturing and disseminating effective workflows remains a challenge, as knowledge about working with agents is scattered across personal configurations, Slack threads, and wikis. One engineer asks, “How can I capture his behavior patterns and spread them across the team?” This fragmentation leads to inefficiencies, as agents start every task cold, and team members repeatedly explain the same conventions. Output quality, in this context, depends more on who is prompting than on agreed-upon team standards. ## Bottlenecks in Team Workflow and Coordination Faster developers don’t necessarily translate to a faster team, as bottlenecks in workflow and coordination can significantly slow progress. According to one product manager, with one product manager per five or six developers, the developers are getting through the work so fast that getting backlog items into a state where they’re ready to be picked up by a person or an agent is challenging . Additionally, when teams automate on their own without coordination, it can lead to fragmentation. As noted by a mobility company, the backend team once asked, ‘Can you give us a key in CI inside GitHub and a secret to auto-generate tests?’ . Finally, if verifying the output costs more than the savings it generated, the team has actually regressed . These examples highlight the importance of alignment and shared practices in maintaining team efficiency. ## Real-World Examples of AI Agent Usage and Limitations Real-world examples highlight both the potential and limitations of AI agents in software development. One agency demonstrated rapid response by turning a vague customer ticket into a ready pull request within 10 minutes, showcasing the speed AI can bring to certain tasks. However, a principal engineer at a gaming company emphasized the gap between experimentation and production readiness: 'Experimenting is one thing, but putting stuff into production – with all the flows, the good and the bad, reviews, and all the things people have to deal with in 2026 – is another thing.' This highlights that AI tools may not yet handle the full complexity of real-world workflows. Additionally, the bottleneck can move upstream, as noted by a cloud consultancy CEO: 'If you make a bad prompt, the codebase is read three times.' These insights reveal that while AI can accelerate certain aspects of development, it does not eliminate the need for human judgment, collaboration, and refinement in complex environments. ## Future Focus Areas for Improving Team Collaboration with AI Faster developers don’t necessarily translate to a faster team, but AI can help bridge this gap by improving collaboration. Future focus areas for teams include shared context and skills, reviews tailored for AI-generated output, and automations that the entire team can own and refine. Air Teams is one example of a platform that enables agentic workflows to be shared across a team, moving away from isolated development on individual laptops. With Air Teams, workflows can be configured once and then run, validated, and improved by the whole team. This approach fosters collective ownership and ensures that AI tools are not just used by individuals but integrated into the team’s broader processes. As teams adopt these practices, the emphasis shifts from individual speed to collective efficiency and shared responsibility. ## Sources and reporting notes - [JetBrains](https://blog.jetbrains.com/air/2026/10/faster-developers-don-t-make-a-faster-team/?ref=stirup.si) Based on published source material. No independent hands-on testing is claimed.