DevOps has come a long way, from simple automation scripts to AI-assisted platforms that understand developer intent.
In this episode, Capital One’s Mihir Vora joins us to unpack how DevOps culture, architecture, and compliance have evolved and where they’re headed next. From the human side of modernization to intent-based engineering, Mahir and Arin explore how today’s engineer can balance innovation, safety, and scale while keeping people and purpose at the center of technology.
Whether you’re a DevOps engineer, architect, or tech leader shaping transformation, this conversation will help you rethink what modern software delivery can look like in the AI era.
Listen to the full episode on Spotify
Listen to the full episode on Apple Podcasts
Watch it on YouTube:
About Our Guest
Name: Mihir Vora
What he does: Mihir is a Senior Distinguished Engineer at Capital One with a passion for empowering teams and driving innovation. A firm believer in “influencing without authority,” Mihir has successfully balanced technical contributions with strategic leadership. With over 15 years of experience in cloud engineering, DevOps, and full-stack development, he has been instrumental in transforming banking through technology. His expertise in optimizing developer experience and accelerating software delivery has led to significant improvements in productivity and efficiency.
Links from the episode: Mihir’s article on intent-based engineering
Key Insights
⚡Culture drives every transformation. The hardest part of modernization isn’t code; it’s culture. Technology aspects are easy compared to the human element of transformation. Mihir explains, “I think more than the code and the cloud piece of it, I think the primary thing is culture and the sentiment behind something. So you can put up architectures, you can put up diagrams, you can put up documents around ‘Hey, here’s why it is better,’ but what is it that you’re going to give incentivizing the people who would benefit from doing it and taking them along the journey? So I think more so than just the technology aspect of it, the culture and the organizational shift is kind of the major part of the transformation.”
⚡Digital transformation never ends. Modern systems are constantly in motion. Success means building flexible frameworks, not clinging to a single solution. Mihir says, “The architecture is not constant; the solution is not constant. You have to keep evolving. Every six months, something’s going to phase out because advanced. And that’s where this whole monolithic services to microservices to containers to whatever is all kinds of evolving aspects of tech, and tomorrow, there might be something else.”
⚡Compliance and innovation can exist. Compliance doesn’t have to slow innovation; it can enable it. By automating controls and embedding risk management into the development process, organizations can move faster and stay secure. As Mihir explains, “It’s a nuance thing to be honest. […] Controls run continuously in the background, and the evidence is captured, things are happening, and then as soon as they find something, they report back. And then that reporting loop could have the developer in the mix, which says, here’s what’s going on, here’s what you need to go fix, and you just treat them as a companion rather than I have to go do this check the box thing just because they kind of nag me. Instead, make it such that the experience is baked in and the developers benefit from it.”
Episode Highlights
Don’t rush your tech transformation.
When it comes to tech transformations, it’s tempting to rush and showcase quick wins, but real, sustainable change comes from a measured approach. Prioritizing small, incremental steps ensures safety, compliance, and a stronger foundation for scaling.
Mihir explains, “The takeaway for me, doing multiple transformations over the years, is you have to start small, you have to ship smaller, prove that it’s working, it’s safer, it’s compliant, it’s tied into the controls, and then start scaling it. And again, the scaling tech basically kind of fits into there. The outcomes are always going to be important, but the journey is equally important to how you got there. I’ve seen teams, engineers, go for the fastest path out to show progress and to kind of get it done faster. I’m kind of not against it, but I feel like we have to balance that with doing it right.”
Developer experience is everything.
A great developer experience isn’t just about tools or processes. It’s about empowering developers to stay focused, creative, and in flow. When you reduce their cognitive load, that’s when innovation thrives.
As Mihir explains, “This is one of my passionate areas, so I can talk about this all day long. Especially at larger organizations, I think the developer experience is key to the advancement of that organization as well. I think of developers as the core asset of the company that helps innovate, bring things faster, evolve the system, so on and so forth, but what developer experience truly means to me is how do I help reduce the cognitive load on the developers? How can I keep them or maximize the flow state where they can head down, focus on doing the things that they should be doing or they want to do, rather than being worried about what happened to this thing and what happened to that thing? Or I have to go look at these five things in order for me to kind of be aware.”
AI empowers DevOps engineers.
As AI continues to transform the way we build and manage software, the role of the DevOps engineer is evolving. Rather than being replaced, engineers are now using AI as a tool to architect, plan, and optimize platforms at scale. Mihir explains, “I think the DevOps engineer basically would use AI as an enabler for building that platform. The platform is not going to build by itself with the context of the enterprise, with the context of everything that needs to happen, so to me, the DevOps engineer’s role shifts a little bit, where they are kind of the quote-unquote architect of the ecosystem, so they can use AI for planning and designing the ecosystem to the best practices. They’ll be using AI to ensure that comprehensive testing is happening for the platform constructs because those platform constructs are now going to impact everyone.”