Docker’s 2025 State of Application Development Report: An In-Depth Analysis
The 2025 Docker State of Application Development Report offers a detailed look into the rapidly changing landscape of software development. This third installment of the report draws insights from over 4,500 developers, engineers, and technology leaders—three times the number of participants from last year. This expanded survey explores the tools, workflows, challenges, and trends shaping the industry today. Key themes from this year’s report include the growing influence of artificial intelligence (AI), the shift towards shared security responsibilities, and ongoing challenges in development despite improvements in tools and organizational culture. With a more diverse respondent base compared to previous surveys, this report delivers a comprehensive view of how modern software is built and how organizations operate.
Key Findings from the 2025 Report
The report highlights several critical findings:
- AI Adoption Across Industries: The survey reveals that 76% of IT and SaaS professionals use AI tools in their work, compared to just 22% across other industries. There is a wide range of AI adoption, with usage varying from as low as 1% to as high as 84% across different sectors.
- Security as a Shared Responsibility: Security is no longer seen as a standalone specialty, especially when vulnerabilities arise. Only 1 in 5 organizations outsource their security needs, and security is a major concern for nearly all respondents, with only 1% indicating it is not a concern for their organization.
- Rise in Container Usage in IT: Container usage in the IT sector has surged to 92%, up from 80% in the 2024 survey. However, adoption lags in other industries, where it stands at 30%. This disparity may be due to IT’s greater reliance on microservice-based architectures, which benefit from the modularity and scalability containers provide.
- Shift to Non-Local Development Environments: A significant shift has occurred, with 64% of developers now using non-local environments as their primary development setup, compared to 36% who still rely on local environments. This marks a reversal from last year’s trend.
- Data Quality as a Bottleneck for AI/ML Applications: Data quality issues are a major hurdle for building AI and machine learning-powered applications. Approximately 26% of AI builders lack confidence in preparing the right datasets or do not trust the data they have.
Developer Productivity, AI, and Security: Key Themes
The report delves into three main areas:
- Developer Productivity: Despite improvements in culture and tools, developers still face challenges. Issues such as delayed pull requests and tasks lacking clear estimates are common friction points in the development process.
- AI’s Impact on Software Development: Contrary to popular belief, AI’s integration into software development is not as pervasive as one might think. While AI tools are making tasks easier, their adoption is uneven across different roles and industries.
- Security as a Collaborative Effort: Security is increasingly seen as a team responsibility rather than a task for a specialized team. This shift emphasizes the importance of integrating security into every stage of the software development lifecycle.
1. Developer Productivity: Opportunities and Obstacles
Developers are thriving in environments with great culture and robust tools. However, persistent challenges remain, such as delayed pull requests and tasks without clear time estimates. The learning landscape is also evolving, with self-guided learning on the rise. A significant 85% of respondents now turn to online courses or certifications, far outpacing traditional learning sources like schools, books, or on-the-job training.
In the IT sector, schools remain a primary learning venue for coding, with 65% of respondents citing them as their top choice. However, online resources are gaining traction, with 63% of IT professionals learning coding skills through these platforms, up from 54% in the previous year.
When it comes to learning preferences, reading documentation remains the top choice, despite the rise of interactive learning methods. Around 29% of developers prefer documentation over videos or side projects. AI tools play a minor role in the learning process, with GitHub Copilot cited by only 13% of respondents as a preferred learning method.
Discovering New Tools and Open Source Contributions
Developers often rely on the insights and opinions of their peers to discover new tools. The main channels for learning about new tools include developer communities, social media, blogs, and recommendations from friends or colleagues. Within the IT industry, blogs have become the top source for discovering new tools, followed closely by developer communities and social media.
In terms of open source contributions, IT professionals are more active, with 48% having contributed in the past year. However, this marks a decrease from 2024, when 59% of respondents reported making contributions. The biggest barrier to contributing to open source is time, cited by 24% of respondents, followed by not knowing where to start and needing guidance on how to contribute.
Tech Stack and Development Environments
The report provides insights into the technology stack landscape, with a focus on application structures, languages, and frameworks. The use of non-local development environments is now the norm, with 64% of developers using these setups as their primary development environment. This shift is driven by the increasing adoption of cloud-based tools and remote development environments.
Internal Developer Portals (IDPs) are still in the early stages of adoption, with only 7% of developers using them. However, usage increases with company size, with 36% of organizations with over 5,000 employees adopting IDPs.
In terms of operating systems, Linux remains the top choice for developers, used by 53%, followed by macOS at 51% and Windows at 47%. The popularity of Python has surged past JavaScript, making it the most used programming language among developers. Framework usage is more evenly distributed, with Spring Boot leading at 19%, followed closely by Angular, Express.js, and Flask at 18% each.
2. AI and its Impact on Software Development
While AI is often touted as a game-changer in software development, its adoption is not uniform across industries. Only 22% of respondents reported using AI tools in their work, with IT/SaaS professionals leading the way at 76%. A significant strategy gap exists, with 73% of tech companies having a clear AI strategy compared to only 16% of non-tech companies.
Despite 59% of respondents viewing AI tools as overhyped, 64% find them useful and say they make their work easier. ChatGPT remains the most-used AI tool, with usage increasing from 46% to 82% compared to the previous year. GitHub Copilot and Google Gemini also saw increased adoption.
Developers across different roles use AI in varying ways. Seasoned developers use AI for documentation and test writing, while DevOps engineers use it for documentation and navigating command-line interface tools. Software developers often rely on AI for research and test automation.
3. Security as a Team Effort
Security is now considered a collective responsibility, with roles blending across organizations. Only 1 in 5 organizations outsource security, and it is a top priority for nearly all respondents. When vulnerabilities arise, it is an all-hands-on-deck situation, with team leads, DevOps professionals, and senior developers all playing a role in addressing security issues.
SonarQube remains the most widely used security tool, although its usage decreased from 24% to 11% due to a more diverse respondent base. Dependabot, Snyk, and AWS Security Hub also saw lower adoption compared to the previous year.
Surprisingly, security does not rank among the top 10 issues holding teams back, with planning and execution activities being bigger challenges. This indicates that security is better integrated into the workflow than ever before.
Conclusion
The 2025 Docker State of Application Development Report paints a picture of a rapidly evolving software landscape characterized by uneven AI adoption, shared security responsibilities, and ongoing challenges in development workflows. While AI continues to gain traction, its integration varies across industries. The shift towards non-local development environments signals a more cloud-native future, and developers are adapting quickly to these changes.
By highlighting these trends, the report not only documents the current state of software development but also charts a path forward. Docker will continue to evolve to meet the needs of modern teams, helping developers navigate change, streamline workflows, and build the future.
Methodology
The 2025 Docker State of Application Development Report was conducted through an online survey by Docker’s User Research Team in the fall of 2024. The survey was distributed widely across various platforms to reach a diverse audience.
For further insights, readers can refer to the original report on the Docker website.
For more Information, Refer to this article.
