Dawnguard Raises $3.3M as Cybersecurity Moves From Detection to Design

Dawnguard has launched its security architecture automation platform to the public, alongside a new $3.3 million pre-seed extension that brings the company’s total funding to more than $6.3 million. The New York and Amsterdam-based startup is positioning itself around a clear thesis: as AI-assisted engineering speeds up how software is designed, written, and deployed, security can no longer wait until systems are already in production.

The new funding includes backing from existing investor BNVT Capital, with new participation from Curiosity VC and eCAPITAL. Dawnguard is also opening a New York City office, marking a broader push beyond its European base as the company moves from design partnerships into general availability.

The Security Problem Is Moving Earlier

For years, cybersecurity has largely been organized around detection, alerting, incident response, and patching. Those functions remain necessary, but they often enter the process after the most important decisions have already been made: how cloud systems are structured, which services are connected, how data flows, what access models are used, and which assumptions are built into the architecture.

That timing matters. A misconfigured database, an overly permissive identity model, or an insecure network pattern is not always a simple bug that can be patched cleanly after launch. In many cases, it reflects architectural intent, or a gap between what security teams thought was being built and what engineering teams actually deployed.

Dawnguard’s platform is designed to close that gap. Rather than only scanning finished environments, the company focuses on turning secure architecture into deployable infrastructure, giving security and engineering teams a shared workspace where designs can be created, reviewed, translated into Infrastructure as Code, and continuously validated against what is running in production.

From Architecture Diagrams to Infrastructure as Code

On its website, Dawnguard describes the product across three core workflows: Design, Discover, and Deploy. The Design layer allows teams to generate cloud architectures from prompts, code, documents, or image references, then refine them on a collaborative canvas. The platform supports AWS, Azure, and Google Cloud, and lets teams define non-functional requirements that can be translated into structured architecture decisions.

That is important because architecture work often lives in static diagrams, slide decks, spreadsheets, or documentation that may become outdated as soon as implementation begins. Dawnguard’s approach is to make the architecture model itself more actionable. Designs can be converted into deployable Infrastructure as Code, helping reduce the distance between what was approved and what is eventually shipped.

For engineers, this turns security from a late-stage review obstacle into something embedded in the build process. Dawnguard says its engineering workflow can generate production-ready Infrastructure as Code aligned with an organization’s environment and requirements, while applying guardrails automatically and reducing security review cycles.

Mapping What Already Exists

The platform is not limited to new systems. Dawnguard’s Discover product connects to existing AWS, Azure, or GCP environments through read-only access, then imports resources and relationships automatically. The goal is to help teams visualize how services, data stores, and infrastructure components are connected across a live cloud environment.

That matters because many organizations do not have a reliable, current map of their cloud estate. Environments change quickly, teams deploy independently, and architecture documentation often trails reality. Dawnguard’s system is intended to surface not only what exists, but also the relationships between services and the security or optimization issues created by those relationships.

The company also emphasizes AI-driven insights that can identify security risks, misconfigurations, and optimization opportunities in context. Instead of treating every issue as an isolated alert, Dawnguard ties insights back to the organization’s requirements, standards, and guardrails.

A Different Kind of Shift Left

“Shift left” has been a common phrase in cybersecurity for years, usually meaning that security checks should happen earlier in the development lifecycle. Dawnguard’s version goes further upstream. The company is not only trying to test code earlier; it is trying to validate the system design before the code or cloud configuration hardens into reality.

Its security workflow allows organizations to translate policies, standards, and regulatory requirements into guardrails that can be applied across AWS, Azure, and GCP. Architectures can then be evaluated against those guardrails early in the design process, with security and engineering teams collaborating through shared review workflows.

This is where Dawnguard’s timing becomes relevant. AI coding tools are making it easier for teams to generate software quickly, but speed can also increase the risk that insecure patterns are created before governance catches up. If engineering work becomes more autonomous, security processes that depend on manual review queues may become increasingly difficult to scale.

Built by Cybersecurity Veterans

Dawnguard was founded by Mahdi Abdulrazak, CEO, and Kim van Lavieren, CTO, with a team that includes cybersecurity and technology experience from major enterprise, cloud, and military environments. The company says its mission is to redefine “true shift left security” by making security part of system design rather than something added after deployment.

That background helps explain the company’s focus. Dawnguard is not presenting cloud security as only a scanning problem or a compliance reporting problem. It is treating architecture as the control layer where security, compliance, resilience, performance, sustainability, and cost decisions increasingly converge.

The company’s website also notes that Dawnguard is ISO 27001 aligned, with ISO 42001 and SOC 2 listed as coming soon.

The Bigger Shift Toward Design-Focused Security

Dawnguard’s launch reflects a broader change in cybersecurity. Enterprises are not abandoning detection and response, but those tools are no longer enough on their own. As cloud systems become more complex and AI changes how software is built, more risk is being created before traditional security tools have anything to scan.

That creates room for a design-focused security layer: one that helps teams model systems earlier, encode requirements into architecture, generate infrastructure from approved designs, and continuously check whether production still matches intent.

The challenge for Dawnguard will be execution. Security teams already face crowded tool stacks, and engineering teams are wary of platforms that slow them down. To become durable infrastructure, Dawnguard will need to prove that secure-by-design workflows can reduce rework, improve visibility, and help teams ship faster without turning architecture review into another bureaucratic gate.

Still, the direction is notable. If AI accelerates software creation, the next security frontier may not be a faster alert console. It may be the ability to prevent flawed systems from being built in the first place.