This article discusses key themes from the DevOps Institute’s annual Upskilling IT (formerly “Upskilling the Enterprise”) survey, originally covered on devops.com. The analysis and commentary below are editorial in nature.
Overview: What the DevOps Institute Upskilling Survey Covers
Each year, the DevOps Institute publishes its Upskilling IT survey to benchmark how organizations are structuring their DevOps teams, what skills are considered essential, and where skill gaps remain. The survey draws responses from practitioners and leaders across the industry, making it one of the more comprehensive snapshots of where DevOps adoption actually stands — not just where vendors would like it to be.
Three themes from the report are especially worth examining in depth: team organization and its relationship to delivery speed, the tension between technical and soft skills, and the continuing dominance of CI/CD as foundational competencies.
Team Organization and Conway’s Law
One area the survey explores is how DevOps teams are structured and how that structure correlates with delivery speed and overall success. This is well-trodden territory — most practitioners have encountered Conway’s Law, which observes that a system’s architecture tends to mirror the communication structure of the organization that built it.
What is more interesting than reciting the law is understanding why certain organizational models work and under what conditions. Structures that reduce hand-off friction — such as stream-aligned teams with embedded operations capability — tend to outperform those that maintain hard silos between development, QA, and operations. The survey’s findings reinforce this, pointing to team organization as a meaningful lever for improving throughput and reliability.
A recurring challenge, however, is that no single organizational model is universally correct. What works for a 50-person software company can be counterproductive in a highly regulated 5,000-person enterprise. Practitioners should treat survey findings as directional guidance rather than prescriptive blueprints.
The Soft Skills Debate: Technical Depth vs. Interpersonal Effectiveness
The survey addresses the role of soft skills in DevOps roles and characterizes many modern DevOps positions as inherently hybrid — requiring both strong technical capability and effective collaboration. This surfaces a genuine tension that engineering managers navigate constantly.
The practical reality in most organizations is this: technical excellence and interpersonal polish rarely appear in equal measure in the same person. A specialist who produces outstanding work but struggles in open forums can still be enormously valuable — provided the team structure accommodates that reality. Good management often means building the organizational scaffolding that lets deep technical contributors do their best work without forcing them into roles where their communication style becomes a liability.
The survey reflects this nuance: soft skills ranked third among the most important skills identified by respondents. That is still a high ranking, and it signals that interpersonal effectiveness genuinely matters in a DevOps context — particularly for roles that span team boundaries, manage incidents collaboratively, or interface with business stakeholders. But it also confirms that technical competence remains the primary qualification. Soft skills reduce friction; they are not a substitute for the ability to actually build, operate, and secure systems.
The “E-Shaped” Professional: A Useful but Imperfect Model
The DevOps Institute uses the concept of the “E-shaped” human to describe an ideal profile for DevOps practitioners. The model extends the familiar T-shaped concept (broad general knowledge, deep expertise in one area) by adding additional horizontal bars — representing multiple areas of meaningful depth, not just one.
The model is a useful thinking tool during hiring and career planning. It encourages engineers to develop competency in adjacent disciplines — security, observability, infrastructure-as-code — rather than remaining narrowly specialized. It also gives organizations a framework for identifying skill gaps across a team.
The limitation of any such diagram, however, is that abstraction loses detail. Real skill profiles are uneven and context-dependent. An E-shaped diagram cannot capture the difference between someone who has passed a certification and someone who has operated production systems at scale. Managers should use the model as a starting point for conversation, not as a scoring rubric.
CI/CD Remains the Baseline Competency
Perhaps the least surprising finding — but still worth stating clearly — is that continuous integration and continuous delivery (CI/CD) are now considered entry-level requirements for DevOps roles. If you are hiring for a DevOps position or assessing your own readiness, fluency with CI/CD pipelines is table stakes, not a differentiator.
This makes sense historically. CI/CD tooling matured early relative to other DevOps disciplines, and the practices are now well-documented, widely taught, and extensively tooled. Most organizations have some form of CI/CD in place, even if implementation quality varies.
What the survey also implies — and what practical experience confirms — is that CI/CD alone is not sufficient for high-performing delivery. A fast pipeline that delivers into fragile deployment infrastructure, or that lacks meaningful security gates, or that produces artifacts without actionable monitoring, is only marginally better than no pipeline at all. The disciplines that surround CI/CD — deployment automation, security scanning, observability, and incident response — are growing in importance and will continue to do so. Expect future editions of this survey to reflect those areas climbing higher on the required-skills list.
Practical Takeaways for Engineering Teams
- Audit your team’s skill profile honestly. The E-shaped model is a reasonable starting framework. Identify where your team has genuine depth vs. surface familiarity.
- Design org structure before job descriptions. How your teams are structured will shape what skills matter and how people need to collaborate. Don’t hire for skills your org structure won’t let people use.
- Invest in soft-skills training, but keep it proportional. Communication, incident facilitation, and stakeholder management all improve with deliberate practice. Training in these areas returns value — but not at the expense of technical development time.
- Treat CI/CD fluency as a baseline, not a selling point. If your organization is still treating CI/CD implementation as a major initiative, that is a signal to accelerate — not a reason to deprioritize the disciplines that build on top of it.
- Read the full report. The themes discussed here are a subset of a much larger dataset. The complete survey is worth reviewing to understand where peer organizations are on their DevOps journey and to identify where your own gaps might be less obvious than you think.
Final Thoughts
The DevOps Institute’s Upskilling IT survey is one of the more grounded annual benchmarks available to practitioners. It avoids the vendor-driven optimism that inflates many industry reports and surfaces findings that reflect the actual, uneven state of DevOps adoption across enterprises of different sizes and maturities.
The core message across all its findings is consistent: DevOps is fundamentally a technical discipline, executed by people who must also work effectively with other people. Getting that balance right — in hiring, in training, and in team design — is ongoing work, not a one-time initiative.