AI for engineers: what it covers and where it cannot be trusted

AI for engineers: what it covers and where it cannot be trusted

For an engineer the cost of an error is high: a wrong calculation or a missed contradiction surfaces during installation. So what is needed here is not the advice "give it a try" but a boundary. Below: what is covered reliably, what cannot be trusted, a verification protocol taking minutes, and what to do if the tool is not officially permitted at work.

For an engineer the cost of an error is high: a wrong calculation or a missed contradiction surfaces during installation. So what is needed here is not the advice "give it a try" but a boundary. Below: what is covered reliably, what cannot be trusted, a verification protocol taking minutes, and what to do if the tool is not officially permitted at work.

The profession where a mistake costs the most

Among those who answered my survey there are engineers — including from industry and the financial sector. And their situation differs from everyone else's.

They need the tool, and the cost of an error is high. Not "the text came out badly" but a wrong calculation, a missed contradiction in a specification, an error in documentation that surfaces during installation.

So the general advice of "try it, you will like it" does not work here. You need to understand what can be handed over and what cannot, and by what protocol to verify.

Let us go through it by type of task.

What is covered confidently

They all share one feature: you supply the material rather than the model retrieving it from memory. That is the most reliable mode of working there is.

Analysing supplied documents

A specification, a brief, a procedure, someone else's design. You attach the document and ask for it to be checked against specific points.

What actually gets found: internal contradictions, missing items, discrepancies between sections, mismatches between text and tables.

Why this works reliably: the model is not inventing standards — it is comparing what you gave it against itself.

Completeness

Checking against a list: is everything there that should be. The work is mechanical, a person tires and misses things, a machine does not.

Drafts of standard documents

Briefs, memos, process descriptions, instructions. To your template and your inputs.

The saving here is not in quality but in the blank page ceasing to be an obstacle.

Translating the complex into plain language

Explaining a technical decision to a client, an adjacent department or management. One of an engineer's most frequent and least favourite tasks.

Structuring raw material

A meeting transcript into outcomes with owners and deadlines. Notes into a coherent document.

Finding contradictions between versions

Two revisions of a document — what changed and whether it broke something in another section.

Where it cannot be trusted

The honest part, and it matters more than the previous one.

Standards, norms, document numbers. A model names non-existent ones as confidently as real ones. An error looks identical to a correct answer — there will be no signal saying "I am unsure here".

Calculations. The arithmetic comes out, but the choice of formula, coefficients and assumptions is a zone where an error is invisible from the look of the result. A figure looks like a figure.

Anything that was not in the material supplied. If the model is drawing on memory, treat it as inventing.

Default values. Grades, sections, tolerances that are "usually like this". Usually does not mean in your case.

Decisions with consequences. Approval, signature, responsibility — a human.

The verification protocol

Five steps, and it takes minutes rather than hours.

Step 1. Split the answer into two types. Verifiable statements — numbers, values, references, designations. And reasoning — logic, structure, wording. The second you judge with your expertise; the first is always checked against the source.

Step 2. Ask for sources in the request. "For each statement, state what it rests on; where there is no source, say it is an assumption." You get a list of what needs checking instead of continuous text.

Step 3. Set the material frame. The phrasing obligatory for this profession: "rely only on the attached document; if the answer is not in it, say the data is insufficient".

Step 4. Verify against the primary source rather than by asking again. Asking the same model "are you sure?" is pointless: it will either agree or reverse itself, and neither means anything.

Step 5. Responsibility lies with whoever signs. If the document goes out under your name, you answer for every value regardless of who found it.

How much this actually saves

The counting has to be honest, because "checking takes longer than doing it" is the main objection.

Analysing a specification used to take a day. With a model — about twenty minutes plus verification. What has to be checked is not the whole text but the list of verifiable statements: usually five to fifteen items. A minute each is a quarter of an hour.

Total: half an hour against a day. Even if verification takes an hour, the ratio does not change qualitatively.

The saving disappears in one case — when a document contains hundreds of verifiable values. That is a sign the task was chosen wrongly: something like that is not handed over whole but broken up.

What to do if it is not accepted at work

A separate situation, and one engineers meet often: a large organisation, strict procedures, the tool not officially permitted.

Concealment is a bad option. An error in work done with a tool nobody knows about becomes your personal problem twice over: the error plus the concealment.

A workable order:

Start with your own tasks — the ones only you see. Drafts, structuring notes, preparing questions.

Gather figures over two or three weeks: how long it used to take, how long it takes now.

Go to your manager with a finished result rather than a request for permission. Not "may I try this" but "here is the report, I assembled it differently and it took forty minutes instead of four hours, I checked it against sources, here is the verification".

And firmly: nothing sensitive goes out. Internal documents, restricted data, personal information. If in doubt, treat it as forbidden.

For organisations with data requirements there is a separate route — a model deployed inside the perimeter. Then the information does not leave the building at all.

How an engineer earns from this

First — speed for yourself. The freed time goes to work nobody previously got round to.

Second — technical documentation as a service. Small companies are bad at it: either it does not get written, or it gets written by an engineer who hates it. Analysing, structuring and preparing documents is work people buy.

Third — auditing a client's documents. Contradictions and omissions found are a specific, verifiable result. And it is also the best way in to a client: you arrive not with an offer of services but with three precise comments on his document.

The order of prices. From the Parselona feed — 5 986 jobs over 4 months: a system or a pipeline $490–1 460, against a median one-off job of around $60.

What the platform offers an engineer

The "Engineers" category — nine assistants: programming mentor, software engineer, bot architect, IT manager, prompting assistants and specialists by discipline.

The "Strategic analytics" category — twelve decision-making methods, including analysing a failure in advance: before a project launch it finds roughly a third more risks than the direct question "what could go wrong".

The legal department — thirteen assistants, if the documents are contractual.

Gen AI, the "File → text" mode — recognising scans and drawing documentation as text, transcribing meetings. $0.15 per operation.

The express course "Perplexity as an investigative tool" — about the method of verification: how to check a fact, a reference or a value in minutes instead of trusting a text. For this profession it matters more than any other. $9, included in Basic.

Cerebrum — if document analysis is regular: you describe the task in plain words, the right models are picked for you, and the price is visible before launch.

Registration is free. Three days of the Basic plan switch on with one button, no card needed.

Where to start today

One action. Take a document you are currently checking by hand and ask for its internal contradictions — discrepancies between sections, between text and tables, between versions.

Not standards. Not calculations. Contradictions specifically, inside what you attached.

That is the safest task there is — the model has nothing to invent — and the most underrated: a person reading continuously tires and misses exactly this kind of thing.

The second approach: not a chat but a conveyor

Everything above is work in dialogue with an assistant. There is another format, and for volume it suits better.

Cascading prompts in spreadsheets

The course “Earning with ready templates” contains spreadsheets with an integrated model: a large task broken into micro-steps, each living in its own cell and taking the previous result as input. The formula can be dragged down a thousand rows — and a thousand tasks are processed at once. How this works step by step is covered separately: Thirty posts in one launch.

What is available for an engineer:

Eight templates for different tasks. For an engineer the most useful are the HR template for job descriptions and the business strategy template for analysing projects.

The course is included in the Basic plan.

The same tasks when you need images, video and sound: Cerebrum

When a chain needs images, video, sound or calls to external services, a spreadsheet is not enough — Cerebrum takes over: you describe the task in plain words, the right models from Gen AI are picked for you, the price is visible before launch and you pay from your balance by actual use. For a clip with subtitles and sounds there is also the AI SKILLS connector for Claude, on the Full plan.

What of the ready material applies to an engineer

The course “Automator”, lesson "Working with files. Report analysis" — taking cost and technical documentation apart. The model compares sections against each other and finds discrepancies.

The skill “Architecture diagram analysis” (advanced) — turns an image of a flow diagram into structured data: components, links, flow directions. A scan of a scheme becomes a set of objects you can work with.

Open Source: Jan — an offline AI chat on your own computer — an open model runs locally, with no internet. For an engineer in an organisation with data requirements that is the only permissible option: nothing leaves the building.

The "Automator" course — Plus plan; Jan — the description is open to everyone, the repository link from Creator; the skill is advanced (Creator and above, or a one-off $9, or 1000 XP).

Specialist assistants in the catalogue

The "Engineers" category in the catalogue, Basic plan:

Separately, Brainstorm AI: Pre-mortem — what in a project will break first, before it happens on site.

Skills for documentation and diagrams

Note: not one of them calculates for you. That is exactly the boundary this article is about. All four skills are advanced: Creator and above, or a one-off $9, or 1000 XP.

What this changes in substance. The calculation stays with you and with verified tools. The tools above take away what surrounds the calculation: the description, the diagram, the plan, the completeness check. That is usually most of the time.

What to take on the platform for this task

What to read next

How to verify a fact with AI search — the verification method that matters more than any other for this profession.

AI for analysts — a pre-mortem before a project launch.

Local models for sensitive data — if nothing can leave the building.

Transcription and file analysis — recognising scans and documentation.

Best open LLMs 2026 — the full guide to open models.

Bringing AI into your department when you are not the boss — what to do when the decision is not yours.

Also on AI SKILLS