Resources on AI automation, in plain language.
These are the guides, comparisons, templates, and tools Aldenebai uses on the Automation Audit and in the build itself: what the terms mean, how the systems compare to the alternatives, how they wire into the tools you already run, and how to put a number on the work you want gone. Everything here is written by Paul Rahme from production experience, and every page says where AI is not the answer yet.
All resources
The Automation Map
The blank map filled in on the audit: list every repetitive job, score it by volume, time, error cost, and risk, rank it, write the honest not-yet.
Open the template →Support automation ROI calculator
Tickets a month, minutes a ticket, loaded hourly cost, expected auto-resolve rate. Hours and money back, with the assumptions shown.
Run the numbers →QA automation ROI calculator
Releases a month, hours per manual regression pass, hourly cost, expected reduction. The QA hours and money a pipeline would return.
Run the numbers →Zendesk AI agents vs a custom agent
For Zendesk teams: what the built-in feature does, what a custom agent inside Zendesk adds, and when built-in is the right call.
Compare →How to automate customer support with AI
Seven steps from a queue to an agent that resolves tickets: measure, fix the docs, choose resolution over deflection, set the allowed actions, wire, pilot, widen.
Read the guide →How to automate QA testing with AI
Six steps from a manual regression cycle to a suite that generates, runs, and heals its own tests, with the two steps teams skip.
Read the guide →Human-in-the-loop AI automation
The four places to put a person in an automated system, how each production system does it, and a checklist for vendor meetings.
Read the guide →AI automation consultant vs agency
Who does the work, who is accountable, how the cost models differ, and when an agency is the right call.
Compare →The words on this site, defined
Confidence threshold, human-in-the-loop, kill switch, self-healing test, build gate, scale model, pilot. One paragraph each, no jargon about the jargon.
Read the glossary →AI support agent vs chatbot
One deflects, one resolves. Where each lives, what each is measured on, how each fails, and when a chatbot is the right answer.
Compare →n8n vs Zapier vs Make
Which orchestrator for which job, from someone who runs production AI pipelines on n8n and still uses Zapier where it fits.
Compare →AI test automation vs scripted suites
Hand-written Playwright and Cypress suites against a pipeline that writes and heals its own tests, by someone who has built both.
Compare →AI support agent for Zendesk on AWS Bedrock
How the agent is wired into Zendesk, why the model calls go through your own AWS account, and what escalation looks like to your team.
Read the guide →Linear to GitHub PR automation with n8n
The production pipeline, stage by stage: labels, triage, one-at-a-time dispatch, the coding worker, the build gate, the watchdog.
Read the guide →AI automation for marina and dealer software
The two verticals with systems in production: what repeats in marina management and dealer networks, and what a machine already handles.
Read the industry page →How these get used on the audit
The sequence is the same every time. Before the call, the ROI calculator turns four numbers you already know into a rough figure, so nobody arrives arguing about whether the queue is worth touching. On the call, the Automation Map template is the working document: every repetitive job listed, scored by volume, time, error cost, and reversibility, then ranked.
Afterwards you get the filled-in map. The glossary is there so its terms mean one thing each, the human-in-the-loop guide is what its "where the person stays" column refers to, and the comparisons exist so you can check the recommendation against the alternatives instead of taking it on trust.
Why these are free, and unbranded
Because a map you can act on without Aldenebai is the only honest version of the offer. Everything here is written to be used by your own team, quoted in your own documents, and taken to another vendor if that turns out to be the right call. There is no email gate, no sign-up, and no drip sequence attached to any of it.
They are also the fastest way to judge whether the engineer behind them knows the work. That is the point: read the pipeline written out stage by stage and decide for yourself before booking anything.
Questions about these resources
Do I need to give an email address to use these?
No. Every page, template, and tool here is readable and usable with no sign-up, no email gate, and no download form. The ROI calculator runs entirely in your browser and sends nothing anywhere.
Which resource should I start with?
If you have a support queue, start with the ROI calculator, then the seven-step guide. If you are deciding what to automate at all, start with the Automation Map template. If you are evaluating a vendor, take the human-in-the-loop checklist into the meeting.
Who wrote these?
Paul Rahme, the engineer who built the systems described on this site, from production experience rather than research. Where something has not been run in production, the page says so.
How do these get used on the Automation Audit?
The calculator sets expectations before the call, the template is the working document during it, and the filled-in map arrives within 48 hours afterwards. The glossary and the guides exist so every term in that map means one specific thing.
Reading is free. So is the call.
The Automation Audit applies all of this to your company in 30 minutes, and you leave with a written Automation Map.