Skip to content

How it works

Built like infrastructure, not a chatbot.

For the technically curious: here’s the architecture underneath PetroBrain — where it runs, how it grounds its answers, where the numbers actually come from, and the hard boundaries it operates within.

The two-tier model

Knowledge in the cloud. Your operations behind your firewall.

PetroBrain splits cleanly in two. General knowledge and public data live in the cloud tier. Everything proprietary — your documents, your historian, your operational data, and the calculation engine — stays on-prem, inside your tenant, behind your OT firewall. Only scoped context and cited answers cross the boundary, and nothing crosses back as control.

PetroBrain two-tier architectureA cloud knowledge tier containing the reasoning layer, general engineering knowledge, and public market data, separated by an OT firewall from an on-premises operational tier inside the customer's tenant. The operational tier holds the document RAG index, a read-only historian replica, the deterministic calculation engine, and operational data that never leaves. Only scoped context and cited answers cross the firewall; there is no control path back into OT.CLOUD KNOWLEDGE TIERoff-premisesReasoning layerthe LLM — reasons,cites, and orchestrates.Never does the math.Engineering knowledgestandards, codes, anddomain expertise —domain-locked.Public & market dataEIA · OPEC · IEA ·rig counts · satellitemethane.OT FIREWALLscoped context ↑cited answers ↓ON-PREM OPERATIONAL TIERinside your tenantYour documents → RAG indexSOPs · P&IDs · well files · daily reports.Answers are grounded in these.Historian / SCADAread-only replica. PetroBrain reads;it never writes back to control.Deterministic calculation engineevery number is computed here —not generated by the model.Operational dataproduction, maintenance, incidents —stays in your tenant. Never leaves.
Only scoped context and cited answers cross the firewall — read-only, with no control path back into OT.

Cloud knowledge tier

The reasoning layer, general engineering knowledge, and license-clean public and market data. This is the part that scales and improves — without ever holding your operational data.

On-prem operational tier

Your document index, a read-only historian replica, the calculation engine, and your operational data — all inside your boundary. Read-only into OT, with no path to actuate anything.

Grounding & calculation

Answers grounded in your documents. Numbers from an engine, not a guess.

RAG

Grounded in your own documents

PetroBrain retrieves from your actual SOPs, P&IDs, well files and reports, then answers from what it found — and cites it. It isn’t recalling the internet; it’s reading your controlling documents, so the answer reflects your plant, not a generic one.

Deterministic

The model never does the math

Language models are good at language and unreliable at arithmetic. So PetroBrain doesn’t let the model compute. It selects the method; a separate deterministic engine produces the number. Same inputs, same answer, every time — and checkable.

01

Model selects the method

The reasoning layer picks the right formula or procedure and explains why — grounded in your documents and the relevant standard.

02

Engine computes the number

The deterministic calculation engine does the arithmetic. The language model never produces the figure itself.

03

Answer returns with its sources

You get the result, the working, and a citation to the document or standard it came from — ready to verify.

model: select methodengine: compute( inputs )answer: result + citation

Safety posture

A human is always in the loop — by design, not by policy.

The architecture assumes a person makes the call. That assumption is built into how answers are framed and what the system is allowed to do.

It proposes; you dispose

PetroBrain surfaces the reasoning and the recommendation. The decision — and the accountability — stay with the competent person.

Safety-critical = explicit verify

Anything that touches safety carries an instruction to confirm against the controlling document and a qualified person before acting.

Traceable by design

Every answer is reconstructable: the sources it used and the figures it cited, so a reviewer can audit the path, not just the conclusion.

The boundaries

Says no on purpose

What PetroBrain does not do.

The limits are the point. A system you can trust in a control room is defined as much by what it refuses to do as by what it can.

Never actuates equipment

There is no control path from PetroBrain into your OT. It reads a historian replica; it cannot write a setpoint, open a valve, or move a single piece of plant.

Never auto-sends or auto-files

It drafts; it does not dispatch. No report is submitted, no message is sent, and no action is taken on your behalf without a person choosing to do it.

Never answers outside the domain

It’s domain-locked to oil & gas. Ask it something off-topic and it declines, rather than guessing in a field where a confident wrong answer is dangerous.

Never invents the math

Numbers come from the deterministic engine, not the model. If it can’t compute or source a figure, it tells you — it doesn’t fabricate one.

See the architecture in action.

We’ll walk your engineering and IT teams through exactly how it grounds, computes, and stays inside the lines — on your stack.