Ingest
Drop in the documents you already have — PDFs, slides, screenshots, SOPs. OpsCodex reads them in.
BPMN governance for industrial enterprises
From documents to governed BPMN — fast, and auditable.
AI drafts the process from your existing documents. Your people review and approve it. Nothing ships without sign-off.
On-premise. Your data never leaves your environment.
The problem
Process knowledge is scattered. Turning it into trustworthy BPMN is slow.
- Knowledge lives in PDFs, slide decks, screenshots and SOPs — never in one modellable place.
- Hand-modelling each process is slow, and the result depends on who drew it.
- Notation drifts between teams, so diagrams aren't comparable or reusable.
- When an auditor asks who approved this, and when? — the trail is fuzzy.
How it works
Four steps from raw document to approved, exportable process.
AI does the heavy lifting on the draft. Governance is built into the path — not bolted on after.
AI-assisted draft
The AI proposes a BPMN draft with surgical, editable suggestions. You refine the model directly — it's a starting point, not a black box.
Six-eyes governance
An explicitly assigned reviewer and approver sign off, with exception handling and an immutable audit log at every step.
Export
Only human-approved processes export as OMG-conformant BPMN XML — ready for your tooling, with the approval on record.
Governance
AI proposes. Humans dispose.
Every process moves through a mandatory Author → Reviewer → Approver split. The roles are assigned, the sign-off is explicit, and the record can't be quietly edited later.
- Separation of dutiesThe author can't approve their own work — the six-eyes split is enforced, not optional.
- Append-only auditEvery state change is written to an immutable audit log. Nothing is deleted or rewritten.
- Defensible by designWhen an auditor asks who approved what and when, the answer is one query away.
Approval chain — assigned, explicit
Author
drafts the model
Reviewer
checks & comments
Approver
signs off
2026-06-08 09:14 draft.created author=k.bauer 2026-06-08 11:02 review.requested → a.singh 2026-06-08 14:37 review.approved a.singh 2026-06-09 08:20 approval.granted m.weber 🔒 2026-06-09 08:20 export.bpmn_xml sha256:9f3c…
Security & trust
Built for data that has to stay in-house.
OpsCodex runs inside your perimeter, not on someone else's multi-tenant cloud.
On-premise deployment
Runs inside your own environment — not public SaaS. Your processes stay within your perimeter.
Your data never leaves
Documents and models are processed in-tenant. AI providers are configurable, so you control where inference runs.
Role-based access
Granular roles map to the governance workflow. People see and do exactly what their role allows — nothing more.
Append-only audit
Audit events are write-once at the database level. The trail can be read, never tampered with.
Human sign-off required
AI output is a proposal. Nothing exports until a human approver has signed — there is no fully-automated path.
OMG-conformant output
Exports are standards-compliant BPMN XML — portable to your existing modelling and execution tools.
Outcomes
Speed without losing control.
Document → draft, faster
Start from an AI draft instead of a blank canvas. Modelling time collapses from days to an afternoon.
Consistent modelling
Naming templates and shared notation mean every process looks — and reads — the same way.
Defensible approvals
An immutable, queryable trail of who approved what and when, ready for any audit.
A governed path from scattered documents to processes people actually trust.
Built for industrial & regulated enterprises
See OpsCodex on your own processes.
We'll walk you through ingest, AI-assisted drafting and the six-eyes governance workflow — using a process from your world.
Request a demo