--- title: Dutch Security Architecture Startup Dawnguard Opens Its Platform and a New York Office description: Dawnguard has moved its security architecture platform to general availability, raised its pre-seed to $6.3 million and opened a New York office. author: Darie Nani (Editor-in-Chief) date: 2026-07-01T09:03:01.387Z updated: 2026-07-01T09:03:01.398Z canonical: https://www.sovereignmagazine.com/article/dawnguard-opens-platform-new-york-office image: https://cdn.nanimediahouse.com/dawnguard-launch.webp categories: Artificial Intelligence, Science & Tech, Startups content_type: Spotlight region: New York City, Netherlands publication: Sovereign Magazine about: - type: Organization name: Dawnguard description: "Dawnguard’s mission is to redefine cybersecurity with a platform that enables true shift-left security , from day zero to day 10,000. For more information please visit https://dawnguard.ai/" url: https://dawnguard.ai/ --- Dawnguard, a cybersecurity company founded in Amsterdam, has opened its security architecture platform to any customer and set up an office in New York to sell it. The move from private trials to general availability comes with a further $3.3 million in funding, taking its pre-seed round to $6.3 million, and a push into the United States that the company says its enterprise customers are asking for. The company [launched from stealth earlier this year](https://www.sovereignmagazine.com/article/dawnguard-bets-on-security-by-design-can-dutch-led-cybersecurity-fix-it-s-flaw-at-the-root) with $3 million and a single argument: most breaches trace back to decisions made when a system is designed, not to the tools watching it once it is running. Since then it has worked with more than a dozen enterprise design partners and turned that argument into a product that is now on general sale. ## Why Security Architecture Is Moving to the Design Stage For two decades, cybersecurity has been built around detection and response. Watch a running system, raise an alert when something looks wrong, patch it. That model is under strain. Software is now written and shipped faster, often with the help of AI coding tools, and the same kind of AI is being used to find and exploit weaknesses at machine speed. The security industry has taken to calling this the Mythos era, after [the Anthropic model that showed an AI system could discover software flaws on its own](https://www.sovereignmagazine.com/article/claude-code-security-is-what-cowork-should-have-been). Much of the risk sits in the design. Gartner forecast that through 2025, 99 per cent of cloud security failures would be the customer's fault, most of them down to misconfiguration rather than any flaw in the cloud platform itself. The US government has pushed in the same direction. Its cyber agency, CISA, published a secure by design framework in 2023 and a voluntary pledge in 2024 that more than 250 companies have signed. Dawnguard is selling into that shift. ## How Dawnguard's Platform Works The platform starts at the architecture, before any code is deployed. Teams design a cloud system in a shared workspace, the platform checks that design against [security and compliance rules](https://www.sovereignmagazine.com/article/cybersecurity-solutions-trust-management-platform-vanta-hits-2-45b-valuation), and it then generates production-ready infrastructure as code from the approved version. Once the system is running, it keeps checking that the live environment still matches what was signed off, and flags any drift between the two. 'For twenty years, security was something you added later,' said Mahdi Abdulrazak, chief executive and co-founder. 'Against an attacker running at machine speed, it becomes increasingly indefensible. When probing is continuous and cheap, the only thing that holds is what was designed correctly from the start.' Kim van Lavieren, the chief technology officer and co-founder, was an offensive security specialist in the Royal Netherlands Navy before running security teams at Amazon. He puts the problem in engineering terms. 'Every engineering team understands the gap between what was designed and what ultimately gets deployed,' he said. 'That gap is where risk lives. Security should not exist in documents, spreadsheets, or diagrams. It should exist in the systems themselves.' ## Who Dawnguard Competes With in Cloud Security The category Dawnguard is trying to define sits on top of several that already exist. Threat-modelling tools such as IriusRisk and ThreatModeler, which merged at the end of last year, also assess risk at the design stage, but generally do not produce deployable infrastructure. Scanners such as Checkov and Snyk check infrastructure code for mistakes before it ships. Cloud security posture tools such as Wiz and Palo Alto Networks' Prisma Cloud watch running systems for misconfiguration after the fact. Dawnguard's pitch is to join those steps into one loop: design the system, generate the code, then reconcile the running environment back to the approved design. Whether that is a category of its own or a set of features the larger players absorb is the question its general availability will answer. ## The Investors Backing the Raise The top-up was led by BNVT Capital, a London firm formerly called 9900 Capital, which led Dawnguard's first round. Two new backers joined: Curiosity VC in Amsterdam, which invests in applied-AI software, and eCAPITAL in Germany, which ran the country's first dedicated cybersecurity fund. The three [keep Dawnguard's funding European](https://www.sovereignmagazine.com/article/why-european-boards-now-prioritise-local-security-providers-over-tech-giants) even as its sales effort moves to New York. The company says the money will go on product development, AI-driven architecture tooling and its expansion in North America. Dawnguard is a small company making a large claim, that the cheapest place to fix a security problem is before a system is built. The shift it is selling into, from detection towards design, is real. Whether a startup with $6.3 million can hold its position against much larger incumbents is a separate question. **About Dawnguard** Dawnguard’s mission is to redefine cybersecurity with a platform that enables true shift-left security , from day zero to day 10,000. For more information please visit https://dawnguard.ai/ [Website](https://dawnguard.ai/) ## Further Context **Q: What does secure by design mean in simple terms?** It means building security into a system while it is being designed, rather than adding tools to defend it after it is running. Dawnguard applies the idea to cloud architecture, checking a design against security rules before any of it is deployed. **Q: Is secure by design required by law?** Not in most cases. CISA's secure by design pledge, launched in 2024, is voluntary, and more than 250 companies have signed it. It sets goals rather than legal obligations, though sector-specific rules and government procurement standards increasingly expect similar practices. **Q: What is infrastructure as code, and why does it matter for security?** Infrastructure as code means defining servers, networks and cloud settings in text files that can be version-controlled and deployed automatically, instead of configuring them by hand. It matters for security because a mistake in that code is repeated everywhere it runs, which is why tools check it before deployment. Dawnguard generates this code from an approved design rather than scanning it afterwards. **Q: Is cloud security posture management worth it?** Posture management tools are useful for spotting misconfiguration in systems that are already running, and most large cloud users run one. They are detective by nature: they find problems after deployment. Design-stage tools such as Dawnguard aim to prevent those problems earlier, and mature teams tend to use both rather than choose between them.