Fund updates, portfolio milestones, and partner perspectives.

The developer infrastructure landscape has undergone a radical transformation over the past three years. What was once a relatively simple stack of version control, CI/CD, and cloud hosting has expanded into a complex ecosystem of specialized tools spanning build systems, feature flags, development environments, observability, and increasingly, AI-assisted coding workflows. In this essay, I attempt to map the current landscape, identify the structural forces driving its evolution, and highlight the layers where we see the greatest investment opportunity in the coming decade.

The central thesis is this: developer infrastructure is following the same trajectory that cloud infrastructure followed from 2010 to 2020. We are moving from general-purpose platforms to specialized, composable services. The winners will not be the platforms that try to do everything, but the ones that do one thing extraordinarily well and integrate seamlessly with everything else.

The foundation model revolution has captured the imagination of the technology industry and the broader public. But as an investor focused on applied intelligence, I believe the real value creation opportunity lies not in the models themselves but in the application layer built on top of them. The companies that will generate the most durable value are not the ones training the largest models. They are the ones building the most sophisticated feedback loops between AI capabilities and real-world workflows.

In this essay, I examine the structural dynamics of the applied AI market, explain why vertical applications will outperform horizontal platforms, and share a framework for evaluating applied AI companies that we use internally at Atlas.

The most effective security is the kind you never notice. This might sound counterintuitive in an industry that profits from fear, but after two decades of building and investing in security companies, I am convinced that the next generation of security winners will be defined not by the threats they detect but by the seamlessness with which they operate. The perimeter is dead. Zero trust is maturing. The next frontier is invisible security: systems that enforce trust without adding friction.

I explore the architectural shifts driving this transition, examine the regulatory tailwinds that will accelerate adoption, and make the case that security companies should be measured by how little their users think about security.