InfoQ Culture and Methods Trends Report – 2025

AI Acceleration, Engineering Excellence, and Evolving Team Dynamics

In the annual Culture & Methods Trends Report podcast, InfoQ’s Culture & Methods editorial team, with special guest Charity Majors, explored the shifting landscape of software development culture, tools, and practices. The wide-ranging discussion revealed several key trends shaping how software teams are working in 2025, from the profound impact of AI tools on development practices to evolving perspectives on team performance and observability.

You can also listen to the entire Culture & Methods Trends Report discussion as a podcast with a supporting transcript.

To help navigate current and future trends at InfoQ and our QCon and DevSummit international software development conferences, we make use of the “crossing the chasm” mental model for technology success pioneered by Geoffrey Moore in his book of the same name. We try to identify ideas that fit what Moore referred to as the early market, where “the customer base is made up of technology enthusiasts and visionaries who are looking to get ahead of either an opportunity or a looming problem.”

As we have done for the 2024, 2023, 2022, 2021, 2020, 2019, 2018, and 2017 Culture and Methods trend reports, we present the topic graph for 2025:

For context, this was our topic graph for 2024:

AI’s Double-Edged Impact on Development

The most dominant trend is the transformative impact of AI on development practices. While AI tools have dramatically increased productivity, they’ve also introduced significant quality concerns that teams are still learning to address.

AI adoption has evolved from cautious experimentation to becoming an integral part of development practice. Recent reports, such as The State of DevOps Report, now consider AI-assisted tools like Copilot essential components of modern development workflows. However, these same reports reveal a concerning rise in change failure rates.

“We’ve got larger chunk sizes. We were all, for ages, trying to get down to small chunk sizes. Now you just say, ‘Robot, give me a solution,’ and you get pages of stuff you might not even be able to read.”

Teams are finding that AI tools enable them to accomplish more with less sacrifice of scope, particularly in hackathon environments. One organization’s AI hack week saw teams accomplish everything from implementing dark mode to creating tools that explain database queries or import dashboards from other vendors. The range and depth of what teams accomplished in a short time was remarkable.

However, the panel highlighted concerning statistics about increased failure rates with AI-generated code in production. A recent study showed “300% more code being released and 400% more bugs as a result,” with the ominous caveat: “that they know of.” This echoes the 2024 State of DevOps report, which warns that AI’s speed may lead to larger changelists, violating the DevOps Research and Assessment (DORA) principle of small batches, thereby  increasing instability:

“Drawing from our prior years’ findings, we hypothesize that the fundamental paradigm shift that AI has produced in terms of respondent productivity and code generation speed may have caused the field to forget one of DORA’s most basic principles – the importance of small batch sizes. That is, since AI allows respondents to produce a much greater amount of code in the same amount of time, it is possible, even likely, that changelists are growing in size. DORA has consistently shown that larger changes are slower and more prone to creating instability. 

Considered together, our data suggest that improving the development process does not automatically improve software delivery – at least not without proper adherence to the basics of successful software delivery, like small batch sizes and robust testing mechanisms.”

This reality presents a new challenge: teams are accustomed to debugging by finding the expert who wrote or deeply understands a piece of code, but with AI generation, that designated expert may no longer exist. The industry is only beginning to develop practices for working with and maintaining AI-generated code.

The Haves vs. Have-Nots in Technology Adoption

A clear divide is emerging between organizations that have embraced new technologies and those still hesitant. This technology adoption gap reveals a stark contrast between cutting-edge teams using the wide range of available AI technologies and more conservative organizations restricting access to limited, locked-down tools like Microsoft Copilot.

While some more mature organizations are pushing boundaries, many others are only just figuring out how to work in an agile way. Without a solid foundation of established practices, adding newer technologies and methods becomes increasingly risky.

This divergence is likely to continue growing. As early adopters refine their usage of AI tools and develop guardrails against quality issues, organizations that restrict access may fall further behind in capabilities and expertise.

Team Collaboration in an AI-Enhanced World

While AI can increase individual productivity, the panel expressed concern about its potential impact on team collaboration. High-performing software development fundamentally depends on effective teamwork. Software development is about working together. The key thing to high-performing software development teams is collaboration, and AI can potentially inhibit collaboration.

Engineers are increasingly turning to AI for answers rather than consulting colleagues, potentially undermining the collaborative culture that drives high performance. 

“People are asking much more stuff on AI and then looking at what AI is producing and looking like, ‘Okay, we got the answer right now, so we don’t need to go to other people in the organization anymore.'”

The panel agreed that in the rush to adopt AI, teams must preserve spaces for human collaboration, reflection, and learning. The importance of enabling functions like observability, progressive deployment strategies, and validation in production still require human oversight and collaboration.

The Value of Junior Engineers in an AI-Enhanced Industry

Despite concerns that AI might diminish the role of junior engineers, the panel strongly argued for their continued importance. Industry voices suggesting juniors are no longer needed were challenged:

“I’ve seen a lot of folks out there saying things like, ‘Oh, I’m never going to hire another junior engineer,’ and I think that is so short-sighted. This is an apprenticeship industry. We all learned by learning from other people.”

Far from being replaced by AI, today’s junior engineers are adapting quickly, using AI as a learning accelerator. Junior engineers are using AI to refine their questions before approaching senior colleagues, making their interactions more efficient and focused.

Building teams that value learning and continuous improvement provides a natural home for junior engineers, who stress test a team’s knowledge-sharing capabilities. Organizations dismissing junior engineers risk losing a vital component of long-term team resilience.

High-Performing Teams: Psychological Safety and Reflection

When discussing what makes teams high-performing in 2025, the panel highlighted several enduring principles alongside emerging practices.

Psychological safety remains fundamental: 

“If you want to build high-performance teams, you need to have everybody really involved in the team, everybody contributing the best thing that they individually can do… if you want to make that happen, a key thing is that there should be psychological safety in the team.”

Post-pandemic organizational cultures, combined with economic pressures, have challenged psychological safety in many teams. Many organizations are cutting back on coaching and team enablement roles, leaving managers or promoted engineers to fill these gaps without adequate training or preparation.

From a technical perspective, tight feedback loops are essential. Reduce the amount of time between when people are writing code and when it’s in production. Make that feedback loop as short and as tight possible.

High-performing teams require some slack in the system. Any team that is running at 100% utilization is at a standstill. 

“If you’re running a team at 90% all the time, you’re burning people out, and that takes a long time to recover from.”

Metrics and Improvement Cycles

Engineering metrics emerged as an increasingly important tool for team improvement. Tools to track lean waste are being used to collect data when engineers lose time due to build failures, unclear requirements, or other impediments.

High-performing teams are not afraid of metrics, but metrics should be conversation starters not conversation enders, not evaluation tools. Don’t use metrics to make judgments about people, use them to help identify bottlenecks in the system and then address those bottlenecks.

Regular review cycles looking at metrics like cost, cycle times, throughput, incident counts, and recovery times – alongside product metrics – provide essential feedback for continuous improvement. These reviews don’t need to be time-consuming; even short, regular check-ins deliver value.

The mere act of making time for improvement is significant; looking at performance, what are the metrics telling us, being curious, trying to really understand what’s happening in the team, is the key to enable improvement.

The Observability Cost Crisis

One of the most striking trends is the rapidly escalating costs of observability. Gartner statistics show one customer’s observability costs growing from $50,000 annually in 2009 to $14 million in 2024 – a 40% year-over-year increase for 15 years.

“This is going to put us all out of business if we don’t get a handle on this,” one panelist warned.

Organizations must determine whether observability is a cost to minimize or an investment that pays dividends. Organizations using between 10-20 observability tools are effectively creating a 10-20x multiplier on their observability costs versus their business growth.

However, growing observability costs partly reflect the increasing complexity of systems. The complexity of systems is exploding, curving up upwards faster and faster. Modern systems are so complex and change so rapidly that engineers can’t rely on mental models, the cognitive load is often too high. They need robust observability tools to understand system behavior.

One possible solution may lie in data lake approaches where observability data is stored once and accessed through AI-enabled interfaces. Observability purchasing decisions should be evaluated as investments rather than just costs.

Beyond “Agile” and “DevOps”: What’s Next?

Terms like “Agile” and “DevOps” were barely mentioned in the discussion, suggesting these approaches have become so integrated into industry practice that they’re no longer explicitly labeled.

“The culture and methods area in InfoQ once was really all about things like agile, and we were in that post-agile period. It’s interesting that it hasn’t really been mentioned once in this whole podcast, which sums up the world.”

Another observation: “I feel this way about DevOps. I feel like we are in the twilight of the DevOps movement, not because DevOps is no longer relevant but because it’s now the air that we breathe.”

Looking to what might be next, constantly improving developer experience is important. Platform engineering was highlighted as a promising direction, described as bringing “a product mindset and a design mindset to tools that you’re building for engineering experience.” The recognition that “engineers are people too, and good design might make our developers more productive” represents an important evolution in industry thinking.

The principles from Gene Kim’s Three Ways, focusing on flow, feedback, and continuous learning, provide a useful framework for thinking beyond traditional agile approaches. These principles align with many of the trends discussed, from accelerating development flow to improved observability feedback and creating learning environments.

Conclusion: The Unevenly Distributed Future

The panel concluded with reflections on William Gibson’s famous observation that “the future is already here – it’s just not evenly distributed.” This perfectly captures the current state of the industry, with some teams leveraging cutting-edge tools and practices while others still struggle with basics.

Several positive industry developments were noted, including the emergence of platform engineering and the establishment of staff-plus engineering career paths that provide advancement opportunities beyond management. Senior individual contributors are often the ones creating programs to recruit and mentor junior engineers, recognizing that “nobody has the credibility of a builder to talk about what it is that builders need.”

For teams looking to thrive in this rapidly evolving landscape, the key lessons include:

• Embrace AI tools but develop guardrails and quality practices around their use
• Preserve and strengthen team collaboration
• Invest in junior engineers and create learning environments
• Implement tight feedback loops between code creation and production
• Make observability a strategic investment while managing its growing costs
• Maintain slack in the system for experimentation, reflection, and improvement
• Recognize that practices like DevOps and Agile are now foundations, not differentiators

By focusing on these areas, development teams can navigate the complex challenges of 2025 while building more resilient, high-performing organizations capable of delivering increasingly complex software systems.