The Automation Audit: 30 free minutes, one written map.
The Automation Audit is a free, 30-minute remote call in which Paul Rahme maps the repetitive work in your company and sends a written Automation Map within 48 hours: what to automate first, what it takes, what it returns, and a first estimate of price and timeline for the first system. Yours to keep, whether or not we ever work together.
Need a website, an app, or an inventory system rather than an AI system? Same call, same 30 minutes: bring the brief.
Proof
You walk away with
01 — The Automation Map
A written map of the repetitive work inside your company: which jobs a system can take over, ranked by return. Sent within 48 hours of the call. See the blank template → Want a rough number before the call? Run the support ROI calculator.
02 — The numbers, defined
What each system takes to build, what it saves, what it costs to run, and where AI is not the answer yet. Engineering talk, not a sales script.
03 — The first move
One concrete recommendation you could start this quarter, with or without us, scoped to a slice where the return is measurable from day one.
What one row of the map looks like
Taken from the support agent at the marina-software client. Your map has one of these per repetitive job, ranked by return, plus the rows marked "not yet".
Not worth your 30 minutes if
The map's most useful column is "not yet". Here it is in advance.
The call is with the engineer, not a salesperson
Paul Rahme: AI automation engineer, 6+ years across QA automation and full-stack delivery, six autonomous systems in production at a US marina-software company. The person who maps your work is the person who builds the machine.
Send one line. Get three slots. Pick one.
Pick a time. Done.
Email one line about what you would bring — a repetitive process, or a brief for a website, app, or internal system — plus your timezone. Within 24 hours you get three slots from Paul’s calendar; reply with one and the invite lands. No form, no sequence, no follow-up campaign.
No forms, no back-and-forth emails. Choose a slot that works in your timezone. You'll get the invite and a short note on how to make the 30 minutes count.
- ▸Bring one painful, repetitive process, or one brief. We'll map it live.
- ▸Bring whoever owns that process today: you, your ops lead, or the person doing the work.
- ▸Nobody needs to take notes. The written map lands within 48 hours.
After you book
Paul replies from his own calendar with three slots in your timezone. Reply with one and the invite lands with a two-line prep note: the one process to bring, and who is worth having in the room.
Calendar invite lands with a two-line prep note: the one process to bring, and who's worth having in the room.
We map your repetitive work live: inputs, outputs, edge cases, guardrails. You watch your own process turn into a system spec.
The written Automation Map arrives: ranked opportunities, effort, return, and the "not yet" column. Yours to keep, no strings.
A fixed-scope, fixed-price proposal for the first system, starting as a pilot on a slice of real traffic with a named target. It goes live, and the repetitive work stops reaching your team.
Asked before booking, answered here
Is the Automation Audit really free? What's the catch?
Free, 30 minutes, and the written Automation Map is yours to keep. The catch is transparent: some companies read the map and commission Aldenebai to build the first system. Many just keep the map. Both outcomes are fine.
We have no AI experience. Is it too early for us?
No. The audit is built for teams starting from zero. We talk about your processes in plain language, queues, releases, tickets, and translate them into systems. No AI knowledge needed on your side.
What about our data and confidentiality?
Everything you share stays confidential, and an NDA can be signed before the call. Systems built later run on your infrastructure, under your access controls. Guardrails and audit trails are part of every build.
What does a system cost after the audit?
Every commission is fixed-scope, fixed-price: what the system does, its guardrails, timeline, and price, agreed before any build starts. The map ranks options by return, so you decide with numbers. How pricing works is explained on its own page.
What happens if the first system misses its target?
You stop, and the most it cost you is the pilot's fixed price, agreed before anything was built. You keep the system, its tests, its runbook, and the measurements that showed it missed, and those same numbers usually say whether it was the design or the data. Widening the slice is a separate decision with its own price, so nothing rolls forward automatically.
Who owns the code, and what happens if Paul is unavailable?
You do. Every system is built in your repositories, on your accounts and infrastructure, with documentation and tests, so any competent engineer can run or extend it. There is no lock-in to Aldenebai, and no licence to keep paying. Every system ships with a runbook and a kill switch; switched off, the work routes back to your team exactly as before. If you choose the Operate option, monitoring and response times are part of the agreement.
Do we need an AWS account?
The systems in production run on AWS Bedrock inside the client's own account, and that is where the numbers on this site come from. If you are on Azure or Google Cloud, or have no cloud account yet, say so when you book; the map will say what that changes for your case.
Paul works with the marina-software client. Is Aldenebai a side project?
No. The marina-software client is a contract engagement, and Aldenebai is the practice through which Paul Rahme takes on other companies. Paul runs every audit himself, with a team ready behind him for the builds, and every proposal names its timeline in writing; if a scope does not fit the calendar, the proposal says so before you commit anything, rather than after.
We are a small team with one repetitive job. Is that too small?
No, if the job is frequent enough to matter. One high-volume, low-risk job is exactly how the first pilot is scoped. If the map shows the return would not cover the build, it says so, and you keep the half hour and the map.
How long until a first system is live?
The proposal names the date before anything starts. The first system starts as a scoped pilot: the system, its monitoring, and its tests together, on a slice of real traffic, scoped small enough to show a live result quickly. The timeline depends on the system and your integrations, and it is written into the proposal; for reference, the client's support agent took about a month from kickoff to its first live ticket, and the audit gives you a realistic estimate for your case.
What does a system cost to run each month?
Systems run on your own accounts, so the running cost is the model usage and infrastructure you already pay for, visible on your own bills. A hard monthly ceiling is set in the design so the bill can never surprise you.
What if AI can't actually do our job?
Then the map says so. Part of the audit's value is the not-yet list: where a system would underperform, you'll know before spending anything. That is why the audit is worth 30 minutes even if you never hire us.
How engagements are priced, in detail: how pricing works →