About Synavistra
Who we are, what we ship, and how we engage with you.
Who We Are
Synavistra GmbH is an Austrian AI engineering firm. We apply our AI document pipeline to your domain documents under engagement, deliver every artifact under Apache 2.0, and transition the work to your own infrastructure when we are done. Headquartered in Feldkirch, Vorarlberg, with the EU as our primary jurisdiction.
Our Mission
Together we chart a path forward. Our AI document pipeline turns regulatory PDFs, internal SOPs, contracts, and reference texts into knowledge graphs, structured answers, and printable evidence — applied to your documents, with your team, on terms that do not lock you in.
Our Values
- Apache 2.0, always. Every Synavistra-authored artifact in your engagement is permanently licensed Apache 2.0. No kill switch, no annual relicensing, no gated features.
- Sovereign by design. We run engagements on Synavistra accounts and hand everything over at delivery. You transition to your own infrastructure on your own clock. Air-gap-capable for sensitive data.
- Empirically verifiable. Our public demo (launching Summer 2026) will run entirely in your browser. After the initial AI model download, disconnect your internet and the demo keeps working. The privacy claim is something you prove yourself, not a promise we make.
- Bilateral expertise. We do not replace your domain experts. We apply our methodology to their work; they validate. Your team's judgment stays load-bearing.
- Transparent, opt-in maintenance. Engagements are one-time fees. Maintenance is opt-in time-and-materials, billed monthly with itemized timelogs. No bundled subscriptions, no recurring license keys.
Our Team
Synavistra is led by Nathan Harrold, an experienced digital-transformation leader with a passion for enabling others. Our team combines AI engineering (extracting entities and relationships from documents, building knowledge graphs, training custom AI models, running inference in the browser) with regulatory-domain literacy (EU AI Act, GDPR, sector-specific compliance).
Ready to apply the methodology to your documents?
Whether you are navigating regulatory compliance, building internal knowledge bases from procedural documents, or extracting structured answers from large reference corpora — let us discuss how the methodology fits your scope.
Get in TouchFrequently Asked Questions
Q: What does Synavistra actually do?
A: We apply an AI document pipeline to your domain documents under a fixed-scope engagement. Output: knowledge graphs, structured answers, printable evidence, and a complete recipe (source code, infrastructure recipes, training stack if applicable) you take home under Apache 2.0.
Q: Where is Synavistra located?
A: Synavistra GmbH is headquartered in Feldkirch, Vorarlberg, Austria. EU jurisdiction is our primary operating context; engagements run on EU-resident infrastructure unless a customer specifies otherwise.
Q: What makes Synavistra different from other AI vendors?
A: Three things. First, every artifact we author for your engagement is permanently Apache 2.0 — no kill switch, no relicensing, no gated features. Second, our public demo (launching Summer 2026) will run in your browser; you can verify the privacy claim yourself by disconnecting your internet after the initial AI model download. Third, engagements have one-time fees with opt-in time-and-materials maintenance — we do not bundle a subscription you cannot get out of.
Q: How do I start a conversation?
A: Email partnerships@synavistra.ai with a brief description of the documents you want analysed and the question you want answered from them. We will reply within two business days with a scoping-call invitation if there is a fit.