DevOps Wisdom from the DOES Stage

The Continuous Discussions (#c9d9) video podcast, produced by Electric Cloud, spent several years hosting hundreds of practitioners and thought leaders from the DevOps community. Among its most notable guests were speakers who had also taken the stage at DevOps Enterprise Summit (DOES) — the annual conference focused on DevOps adoption in large, complex organizations.

IT leaders from organizations including Hearst Business Media, Nationwide Insurance, IBM, Capital One, Target, and Pivotal shared candid accounts of their DevOps transformation journeys: the patterns that worked, the pitfalls they hit, and the cultural shifts that made the biggest difference. What follows is a curated collection of key insights from ten of those conversations.

10 Key Insights from DOES Practitioners

1. Mark Imbriaco: The Benefits of a Mature DevOps Transformation

One of the clearest indicators that a DevOps transformation has reached maturity is a shift in how teams relate to incidents and outages. Rather than scrambling to assign blame, mature organizations treat failures as learning opportunities embedded in a blameless culture. Mark Imbriaco, who has led infrastructure and reliability efforts at large-scale organizations, emphasized that the payoff of sustained DevOps investment shows most clearly when teams can recover faster, communicate more openly, and improve continuously without fear.

2. Topo Pal: Teams Should Own Their Pipeline

A shared, centrally managed pipeline can quickly become a bottleneck and a source of misaligned incentives. Topo Pal’s recommendation is straightforward: the team that ships software should also own and maintain the delivery pipeline that gets it there. This end-to-end ownership creates accountability, reduces handoff delays, and gives engineers direct visibility into where friction exists in the path to production. When pipeline problems surface, the team that feels the pain is also the team best positioned to fix it.

3. Nicole Forsgren: Metrics Are More Than Numbers

Measurement is central to DevOps, but the way organizations collect and interpret metrics matters as much as the metrics themselves. Nicole Forsgren — researcher, author, and co-creator of the DORA metrics framework — has consistently argued that metrics must be tied to outcomes, not just activity. Tracking deployment frequency or mean time to restore is only useful if those numbers inform decisions and drive genuine improvement. Metrics divorced from context can just as easily encourage gaming the system as they can encourage real progress.

4. Carmen DeArdo: Driving Business Value, Not Just Technical Change

DevOps transformations that stay confined to engineering teams rarely achieve lasting organizational change. Carmen DeArdo, who led DevOps efforts at Nationwide Insurance, stressed the importance of connecting technical improvements to the business outcomes that executives and stakeholders care about. Framing DevOps work in terms of faster time to market, reduced risk, or improved customer experience makes it easier to secure executive sponsorship and sustain investment over the long haul. Without that business case, DevOps can stall as a purely technical initiative.

5. Ashish Kuthiala: Collaboration and Empathy as Core DevOps Values

Tooling and automation get significant attention in DevOps conversations, but the cultural dimension is just as critical — and much harder to change. Ashish Kuthiala highlighted collaboration and empathy as foundational attitudes that make cross-functional teams actually work. When developers understand the operational constraints that ops engineers face, and when ops engineers appreciate the delivery pressure developers operate under, the resulting conversations produce better systems and faster feedback loops. Empathy is not soft; it is a prerequisite for effective shared ownership.

6. Paula Thrasher: Specialists Still Have a Place in DevOps Teams

The push toward generalist “you build it, you run it” teams can create a false impression that deep specialization is no longer valued. Paula Thrasher pushed back on that narrative. Specialists — security engineers, database administrators, accessibility experts — bring concentrated expertise that generalists simply cannot replicate. The key is integrating specialists into cross-functional workflows rather than isolating them in separate silos. Done well, the presence of specialists raises the overall capability of the team without creating new bottlenecks.

7. Pauly Comtois: Harnessing Engineer Passion

Engineers tend to care deeply about their craft. Pauly Comtois recognized this passion as one of the most underutilized assets in a DevOps transformation. When organizations channel that energy — by giving engineers meaningful problems to solve, encouraging experimentation, and removing bureaucratic friction — they accelerate change far more effectively than any top-down mandate. The implication for leaders: create the conditions where passion can be productive, rather than managing it out of the system.

8. Ann Marie Fred: Give Developers Greater Responsibility

One recurring pattern in DevOps transformations that stall is over-reliance on gatekeeping processes that limit what developers can do without external approval. Ann Marie Fred, drawing on her experience at IBM, advised organizations to extend greater trust and responsibility to developers — including responsibility for production readiness, monitoring, and on-call participation. Greater responsibility creates greater accountability and, importantly, greater motivation to build quality in from the start rather than pushing quality checks downstream.

9. Steve Mayner: The Power of Transformational Leadership

Technology alone does not drive a DevOps transformation — leadership does. Steve Mayner pointed to transformational leadership as the critical enabler that determines whether DevOps initiatives gain traction or fade after the initial pilot. Transformational leaders communicate a compelling vision, model the cultural behaviors they want to see, and actively work to remove organizational impediments rather than simply delegating change to engineering teams. Without visible executive commitment, DevOps efforts tend to remain localized and fragile.

10. Jayne Groll: Overcoming Language Barriers Between Teams

One of the most underappreciated friction points in enterprise DevOps adoption is the language gap between technical teams and business stakeholders. Jayne Groll — co-founder of the DevOps Institute — observed that development, operations, security, and business teams often describe the same problems and goals using entirely different vocabulary. Building a shared language, or at minimum developing the habit of translating across disciplines, is a practical first step toward alignment. Without it, even well-intentioned cross-functional efforts can collapse into misunderstanding.

Common Themes Across DOES Practitioners

Reading across these ten insights, several patterns emerge that are broadly applicable to any organization pursuing a DevOps transformation:

• Culture before tooling. Empathy, collaboration, blameless culture, and shared ownership consistently appear as prerequisites for sustained improvement — not afterthoughts.
• Ownership drives quality. Whether it is pipeline ownership (Topo Pal) or broader developer responsibility (Ann Marie Fred), teams that own outcomes are more motivated to improve them.
• Metrics must connect to outcomes. Measurement without meaningful interpretation can mislead as easily as inform.
• Leadership is non-negotiable. Multiple speakers pointed to visible, committed leadership as the factor that separates DevOps transformations that scale from those that stall.
• Communication is a technical skill. The ability to bridge language gaps between teams — and between engineering and the business — is as strategically important as any deployment automation.

All speaker quotes and insights in this article were originally shared on the Electric Cloud Continuous Discussions (#c9d9) podcast. The original compilation was published at electric-cloud.com.

For too long, ensuring that code is securely written – and bug free – has been a business afterthought. But there’s been new hope for building security into the development lifecycle, thanks to the rise of DevOps, aka rugged software, says Chris Wysopal, CTO of the application security firm Veracode.

DevOps – a truncation of software development and IT operations – incorporates aspects of agile development, including short sprints – perhaps just two weeks in length – that involve planning, designing and implementing new, working software or additional functionality, backed by having customers embedded into the coding team.

Wysopal is delivering a Feb. 15 briefing at the RSA Conference in San Francisco titled “Your Chance to Get It Right: 5 Keys to Building AppSec into DevOps.”

In an interview with Information Security Media Group, Wysopal notes that DevOps, as well as what’s sometimes now referred to as DevSecOps, improves on agile concepts by adding operations personnel into teams and making developers directly responsible not just for quality control, but also the security of the code they’re writing.

“Just like they fail a build when there’s a functionality problem or a performance problem that’s unacceptable … have them fail the build when there are security defects found that can’t go into production,” he says.

In this (audio link), Wysopal describes:

• The business case for integrating security into the application development lifecycle;
• How DevOps teams operate, and where the security buck stops;
• What resources these teams require to help them more quickly find and fix bugs.

Wysopal is CTO of Veracode, as well as a member of the Black Hat review board. He was previously vice president of research and development at security consultancy @stake, which was acquired by Symantec. He also was one of the original vulnerability researchers at The L0pht, a hacker think tank, where he researched vulnerabilities and wrote security software such as Netcat for Windows and L0phtCrack.

What BestDevOps.com Covers

BestDevOps.com was launched as a dedicated resource for engineers and teams working in the DevOps space. The platform focuses on practical, up-to-date content across several core areas:

• News and Updates: Coverage of developments across DevOps tooling, platforms, and practices.
• Technical Articles: In-depth guides on topics ranging from CI/CD pipeline design to infrastructure-as-code and container orchestration.
• Continuous Delivery: Resources focused on shortening release cycles and improving deployment reliability.
• Cloud Application Development: Content aimed at simplifying how teams build, deploy, and scale applications in cloud environments.
• Events: Information on relevant industry events, meetups, and conferences in the DevOps community.

Why DevOps Requires a Broad Skill Set

One of the defining characteristics of DevOps is that it is inherently cross-functional. A practitioner working in this space typically needs exposure to:

• Programming and scripting — for automation, tooling, and configuration management
• Operations fundamentals — networking, Linux administration, monitoring, and incident response
• Cloud platforms — understanding of services from providers such as AWS, Azure, or Google Cloud
• CI/CD tooling — pipelines built with tools such as Jenkins, GitHub Actions, GitLab CI, or CircleCI
• Agile practices — iterative delivery, sprint planning, and cross-team collaboration
• Test automation — integrating quality checks early and consistently in the pipeline

BestDevOps.com aims to be a single destination where engineers can find reliable information across all of these disciplines without having to piece together resources from dozens of separate sources.

Visit the Platform

Explore news, articles, and guides at BestDevOps.com.