Introducing Zunavi: A Drawing Management Platform for Manufacturers (In Development)

The Problem: Drawings Live Everywhere Except Where They Should

Talk to almost any manufacturer — especially in semiconductor and precision manufacturing — about how they manage engineering drawings, and you’ll hear a version of the same story. The “official” drawing is a PDF on a file share. The latest revision might be in someone’s email. The approval happened over a phone call that never got logged. Someone, somewhere, is working off a printout that’s two revisions old.

This isn’t a people problem. It’s a tooling problem. Most manufacturers are stuck with systems that were never designed for how drawings actually move through an organization — through revisions, through approval routes, through the specs and parts and BOMs they connect to.

What We’re Building

Zunavi (図ナビ) is a drawing management platform we’re building specifically for manufacturers who need to track engineering drawings, specifications, parts, and bills of materials with real version control and real approval workflows — not a shared folder pretending to be a system of record.

The core of what Zunavi handles:

• Drawing ledger and revisions — every drawing tracked as a versioned entity with clear revision history, not files that get silently overwritten
• Version comparison — see exactly what changed between drawing revisions
• Drawing OCR and titleblock parsing — extract structured data (part numbers, revision info) directly from scanned drawings
• AI-powered similar-drawing search — find related or duplicate drawings using vector-based similarity search, useful for avoiding redundant part numbers or spotting reuse opportunities
• Change and approval workflows — drawings move through defined approval routes tied to your organization’s actual structure, not ad-hoc email chains
• Part numbering rules and BOM linkage — specs and drawings connect bidirectionally to the parts and bill-of-materials data that depend on them
• 2D and 3D viewers — view drawings directly in the platform, including Japanese Shift-JIS-encoded files, without needing separate CAD software

Why On-Premise First

Our target customers — semiconductor and precision manufacturers — deal with drawings that are, by definition, confidential. A cloud-first tool that assumes your drawings live on someone else’s servers is a non-starter for a lot of this industry. So Zunavi is being built on-premise-first: your data stays on your infrastructure by default. Cloud deployment exists as an option, available through a feature flag, for teams that want a lighter-weight or demo setup — but it’s not the default, and it’s not required.

Where Things Stand

Zunavi is currently in active development, being built as a clean-room rewrite (NestJS, Prisma, PostgreSQL) — it does not carry forward any code from older internal tooling it’s meant to eventually replace. It is not yet released, and we’re not attaching a release date to it. Multi-tenant support and role-based access control are part of the foundation, since the platform needs to work for both small shops and organizations with more formal approval hierarchies.

Why We’re Talking About It Now

We’d rather be upfront that Zunavi exists and is being built than let it stay invisible until launch. If drawing management, version control for engineering documents, or approval workflows for manufacturing are problems you deal with today, we’d like to hear about it — it directly shapes what we prioritize before release.

More on How We’re Building It

We’ve written in more depth about specific engineering decisions behind Zunavi: why we’re rebuilding the drawing ledger from scratch instead of patching the legacy system, why the platform draws an explicit line between “certain” and “optional” features, and why on-premise is the default deployment mode rather than an enterprise add-on.

---

Working on something similar, or want to be kept in the loop as Zunavi develops? Reach out at [email protected].