Six AI systems in production at a US marina-software company

AI automation consultant:
systems that do the whole job.

Aldenebai is an AI automation consultant and engineering studio led by one engineer, Paul Rahme, with a team ready behind him. It builds guardrailed AI systems that take over whole jobs — support tickets, QA, engineering tickets, phone calls, finance work — and the web platforms, apps, and integrations around them.

Free · you leave with a written Automation Map · no pitch, just engineering talk · built by an engineer who tests what he ships

scale model · simulated traffic · not live data

Scale model of the Aldenebai pipeline: tickets enter, are routed to the support agent, the QA pipeline, or the dev agent, and exit resolved. An animation of the production flow, not live traffic.

MARINA-SOFTWARE CLIENT · SUPPORT AUTO-RESOLVED
~60%
MARINA-SOFTWARE CLIENT · MANUAL QA REMOVED
~80%
MARINA-SOFTWARE CLIENT · TICKET → PULL REQUEST
2× faster
IN PRODUCTION · AI SYSTEMS
6
Scale model above: an animation of the production flow, not live traffic. Numbers rounded from production data at a US marina-software company, 2025–2026 — measurement notes on each case →

Where the work happened

Products and programmes worked on

McKinsey & Company
PwC
Saudi Tourism Authority (Visit Saudi)
Eurisko
Shasha Media
mdgroup

The six AI systems run in production at a US marina-management software company, 2025–2026, whose name is not published here. The rest is the earlier engineering record: Invest Kenya for McKinsey & Company; Visit Saudi for the Saudi Tourism Authority, as Eurisko's QA engineer on the PwC-managed programme; Shasha Media's streaming platform at Eurisko; and a UK pharmacy platform at mdgroup. Every engagement, named →

01 · Replaces the support queue · in production

AI support agent: the queue resolves itself, your team keeps the judgment.

An agent that triages, answers, and closes tickets end to end — guardrailed, auditable, escalating to a human only when confidence is low.

support.agent — scale model

Scale model of the AI support agent: a ticket arrives, is triaged, answered from documentation, and closed; low-confidence tickets escalate to a person.

~60%
of tickets resolved with zero human touch
Seconds
to first response — down from hours
02 · Replaces the manual QA cycle · in production

AI test automation: full regression before every release, nobody running a test.

A pipeline that reads your product, writes the coverage, runs it on schedule, and heals its own failures.

qa.pipeline — scale model

Scale model of the AI test pipeline: generated tests run on a schedule, failures caused by product changes heal themselves, real regressions are reported.

~80%
of manual QA effort removed
0
dedicated manual regression cycles per release
03 · Replaces boilerplate engineering · in production

AI dev agent: a ticket lands, a review-ready pull request comes back.

The machine reads the ticket, scopes the change, writes the code and the tests, survives the build gate, and opens the PR. Your engineers approve instead of building from scratch.

ship.pr — scale model

Scale model of the AI dev agent: a Linear ticket is read, scoped, coded with tests, and opened as a review-ready pull request, a proposed code change your engineers approve.

faster from ticket to review-ready PR
Human
on every merge, always
PICK THE JOB

What can be automated? Pick the job that costs you the most hours.

Every cell is a job a machine can take over. Pick one — get the 10-second verdict. The full map for your company takes one call.

What Aldenebai does, in 75 seconds

Captions on by default, no sound needed
intro.film — 75 seconds

Who builds it

Paul Rahme
BUILT BY AN ENGINEER WHO TESTS WHAT HE SHIPS

Paul Rahme

AI Automation Engineer · Founder, Aldenebai

Six years across QA automation and full-stack delivery, now shipping autonomous AI systems in production at a US marina-software company, where he is the contract AI automation engineer. Aldenebai builds the same systems for other companies. The person who maps your repetitive work is the person who scopes the machine, reviews and tests it, and signs it off. No account managers, no agency chain. You keep the code, the accounts, and the keys.

FROM FIRST CALL TO RUNNING SYSTEM

Four steps. The first one is free.

01 · AUDIT
30 minutes, free
One call with the engineer. We map the repetitive work inside your company, live.
02 · MAP
Written, within 48h
Ranked opportunities, what each takes, what each returns, and the "not yet" column. Yours to keep either way.
03 · PILOT & BUILD
Fixed scope, fixed price
The first system starts as a scoped pilot on a slice of real traffic, with its target, guardrails, tests, timeline, and price agreed before a line is written.
04 · RUN
Guardrailed, 24/7, yours
Live on your infrastructure with a human on the exceptions. The repetitive work stops reaching your team.

How engagements are priced, without the sales script: how pricing works →

Production results: ~60% support auto-resolved, ~80% less manual QA, 2× ticket-to-PR

From production systems at a US marina-software company — the case studies →

~60%
of support resolved autonomously
~80%
of manual QA effort removed
2×
faster ticket-to-PR delivery
24/7
production operation, guardrailed

Rounded from production data at a US marina-software company, 2025–2026; definitions and measurement notes are on each case study.

Before you book: five answers

What does Aldenebai do?

Aldenebai automates repetitive work with AI and builds the software around it. It designs, builds, tests, and runs AI systems that take over whole jobs (support tickets, QA, engineering tickets, phone calls, back-office work), automates workflows across tools such as n8n and HubSpot, and builds web platforms, apps, and inventory systems. Paul Rahme leads the mapping, the building, and the testing, with a team ready behind him.

Is Aldenebai an AI automation agency?

No. Aldenebai is led by one engineer, Paul Rahme, with a team ready behind him and no account managers, hand-offs, or anonymous bench. The person on the audit call is the person who scopes, reviews, and signs off the system. One accountable engineer is harder to get wrong than an agency chain; when a build needs more hands than his, the team builds under his review. The full comparison: AI automation consultant vs agency.

Which AI systems are in production today?

Six, all built by Paul Rahme at a US marina-software company, 2025–2026. Three carry published numbers: the AI support agent (about 60% of tickets resolved with no human touch), AI test automation (about 80% less manual QA), and the AI dev agent (about 2× faster from ticket to pull request). The other three are the autonomous bug fixer, documentation automation, and the voice agent on the support line.

What is the Automation Audit?

A free, 30-minute remote call with Paul Rahme in which the repetitive work in your company is mapped live. Within 48 hours you get the written Automation Map: every job ranked by return and risk, with a "not yet" column. No pitch and no obligation; the map is yours either way.

Who is Aldenebai for?

Companies with a support queue, a release cycle, a backlog, or back-office work that repeats, from vertical software firms to small businesses, and anyone who needs a website, an app, or an inventory system built and tested under one accountable engineer. It is not for teams that want a chat widget: these systems close tickets rather than deflect them.

See what a system would take over in your company.

The Automation Audit is free: 30 minutes, and you leave with a written map of what to automate first — what it takes, what it returns. You keep the map whether or not we ever work together.

Paul runs every audit himself, with a team ready behind him for the build. Three time options within 24 hours, in your timezone. A human runs the call.

Not ready for a call? Start with the blank Automation Map, the same one we fill in together. Open the template →
Audit: free, 30 min →