Automation you can audit. Changes you can defend.
OperiProof proves that network changes did what they were supposed to do. We build the validation, testing and rollback discipline that lets an operator change its production network with evidence instead of hope — and we hand back a signed record of every change.
Most operators cannot answer three questions about their own network with evidence.
Does our documented inventory match what is actually deployed?
When someone makes a production change, what is checked before it goes in?
If a change caused an outage tonight, how long until we knew, and how long to unwind it?
Teams generally know the answers are uncomfortable. What they lack is a defensible picture of how bad it is and what fixing it is worth — which is exactly what makes it impossible to get budget for.
We sell proof, not automation.
The deliverable is a signed, machine-readable evidence bundle demonstrating that a network change did exactly what it was intended to do and nothing else. The automation is how that evidence is produced. The proof is what you keep.
- SHA-256 manifest over every artefact
- Cryptographic signature
- Machine-readable data, with a generated human-readable report
Three named things.
The Assessment
A documented picture of how a bad change reaches production today: source-of-truth reconciliation, change-path map, failure-mode analysis, and a prioritised backlog.
Usable with or without us.
The Implementation
One production workflow automated end to end, with pre-change validation, post-change verification, a tested rollback path, a CI pipeline your team owns, and runbooks.
One workflow at a time. Never a platform rewrite.
The Assurance Retainer
Ongoing operation and extension of the pipeline, new workflows added as agreed, and evidence bundles produced for every change.
Optional.
Your infrastructure. Your credentials. Your repositories.
Automation runs on a runner inside your security boundary, making outbound connections only. There is no inbound path from us into your network, and we never hold a credential that opens it.
Client-specific work product is yours outright. The general framework is licensed to you perpetually, irrevocably and royalty-free for your own internal operations — including if you stop working with us, and including if you hand it to another vendor. You are not buying a dependency.
Regional ISPs and broadband operators
Particularly those absorbing a buildout obligation with milestone and location reporting attached.
Multi-site enterprises
With a network team smaller than the estate deserves.
Data-centre and infrastructure operators
With change volume outpacing process.
Infrastructure-heavy SaaS
Where network change is a production risk.
When a change goes wrong in production, how do you find out — and what does it cost to unwind it?
Thirty minutes, no charge, no deck. If the answer is comfortable, we will say so and there is no assessment to sell.