# vStaub

> Personal technical writing by Thomas Staub, an Infrastructure Architect.

vStaub is Thomas Staub's personal publication covering infrastructure, networking, architecture, AI-assisted workflows, operations, engineering decisions, career development, and related technical systems.

Thomas Staub is the author and publisher of vStaub.

## Author

- [Thomas Staub](https://vstaub.com/about/): About the author of vStaub.
- [LinkedIn](https://www.linkedin.com/in/thomas-staub/): Thomas Staub's professional profile.
- [GitHub](https://github.com/tfstaub): Thomas Staub's GitHub profile.

## Writing

- [Blog](https://vstaub.com/blog/): All published vStaub articles.
- [RSS](https://vstaub.com/rss.xml): vStaub RSS feed.
- [Sitemap](https://vstaub.com/sitemap-index.xml): Published site URLs.

## Content Use

Original articles on vStaub are authored by Thomas Staub unless otherwise stated.

Content may be indexed, referenced, excerpted, and linked with attribution to Thomas Staub and the canonical vStaub URL.

Content is not offered for model training or fine-tuning.

## Articles

- [The Case for Continuous Conformance](https://vstaub.com/blog/the-case-for-continuous-conformance/): Continuous conformance can expose avoidable operational work before it becomes an incident, but the engineering responsibility goes beyond another set of alerts.
- [Your Data Is Part of Your Service](https://vstaub.com/blog/your-data-is-part-your-service/): Keeping infrastructure information useful requires domain expertise, reliable shared platforms, and a way to correct what breaks between them.
- [The Trouble With “Single Source of Truth”](https://vstaub.com/blog/the-trouble-with-single-source-of-truth/): Different systems can be authoritative for different information about the same infrastructure. Finding a discrepancy doesn't automatically change where that authority belongs.
- [What Enterprises Actually Need from Technology Vendors](https://vstaub.com/blog/what-enterprises-actually-need-from-technology-vendors/): Technology vendors add value when they bring context, experience, and honest perspective to the decisions enterprise buyers are actually trying to make.
- [Make Your Career Goals More Obvious](https://vstaub.com/blog/make-your-career-goals-more-obvious/): Doing good work matters, but career direction becomes easier for others to support when you make the work and role you want next explicit.
- [Buying Software Is Still an Engineering Decision](https://vstaub.com/blog/buying-software-is-still-an-engineering-decision/): Choosing software is partly choosing which engineering, operating, and support work an organization is prepared to sustain after the decision.
- [Your Platform Shouldn't Have to Invent the Network](https://vstaub.com/blog/your-platform-shouldnt-have-to-invent-the-network/): Automation works best when platforms make deliberate network architecture easier to consume instead of turning implementation choices into architectural defaults.
- [Five Nines Is Not a Requirement](https://vstaub.com/blog/five-nines-is-not-a-requirement/): Reliability targets only become useful when they reflect the failure the business needs to survive and the cost and complexity it is willing to carry.
- [The Future of Network Engineering Is Bigger Than Networking](https://vstaub.com/blog/the-future-of-network-engineering-is-bigger-than-networking/): Networking fundamentals become more useful, not less, when engineers use them to understand cloud, security, observability, automation, and the systems around the network.
- [Self-Service Networking Should Expose Intent, Not Configuration](https://vstaub.com/blog/self-service-networking-should-expose-intent-not-configuration/): Good network automation asks consumers for the outcome they need, then derives the implementation details the network already knows how to provide.
- [You Should Define Your Segmentation Policy First](https://vstaub.com/blog/you-should-define-your-segmentation-policy-first/): Durable segmentation starts by defining required communication and its consequences before choosing how to enforce it.
- [Network Automation Does Not End at Deployment](https://vstaub.com/blog/network-automation-does-not-end-at-deployment/): Network automation is not complete when the configuration deploys. It also needs evidence that the network, applications, and security controls behaved as intended.
- [What Architects Actually Produce](https://vstaub.com/blog/what-architects-actually-produce/): Architecture produces diagrams and documents, but its more consequential outcome is often the shared understanding that helps an organization make decisions and move work forward.
- [Have You Thought of an Architect as a Lead Engineer? There's More to It Than That.](https://vstaub.com/blog/architect-not-lead-engineer/): Architects get expected to act like lead engineers with a wider scope. The job is usually something else, and the difference tends to show up in small, easy-to-miss moments.
- [AI Can Write the Architecture Document. It Cannot Create Alignment.](https://vstaub.com/blog/ai-can-write-the-architecture-document-it-cannot-create-alignment/): AI can make architecture documents easier to produce, but alignment still depends on shared understanding, visible trade-offs, clear ownership, and people making a decision together.
- [AI-Assisted Network Engineering Is Not Network Automation](https://vstaub.com/blog/ai-assisted-network-engineering-is-not-network-automation/): AI can help network engineers reason, draft, and review, but it does not replace deterministic automation, testing, or the engineering judgment that makes infrastructure changes safe.
- [Relaunching vStaub](https://vstaub.com/blog/relaunching-vstaub/): Why this site is starting small: infrastructure, AI, architecture, career observations, and the useful experiments that keep changing how I think.