Practical AI implementation,
led by strategy.

Solving the right problem before writing any code.

I help businesses find the highest-leverage place to apply AI, then build a working first version within 30 days.

No pitch. We look at one workflow together and you leave with a clear next step, paid or not.

How I work

From a conversation to a working system in weeks, not quarters.

This is how I approach an engagement. Each step stands on its own, so you can stop at any point and still walk away with something useful.

Discovery call
(free, 30 min)

I help you identify two or three workflows where AI could realistically save time or money. You leave with a clear view of what is worth building and what is not, whether we work together or not.

Focused assessment
(1 week, fixed fee)

I go deep on the highest-value workflow, talk to the people doing the work, and come back with a build plan: scope, tools, timeline, and expected outcome. No 40-page reports.

Build and hand over
(typically 2–4 weeks)

I build the system, document it, and train your team to run it. You own what gets built. I stay available for tweaks once it is live.

Start small. Prove value. Expand from there.

AI operating system

The system I run my own business on.

An AI operating system is a layer of AI wrapped around a business. It plugs into the tools and data you already have: calendar, messages, documents, numbers. Part of it runs on a schedule without being asked, so the numbers are pulled, the brief is written and the overdue invoice is flagged before you sit down. The rest you talk to all day, and it answers already knowing the business.

I built one because I run this business alone. Mine writes my morning brief, tracks which invoices are still unpaid, drafts content, and keeps the pipeline in front of me. It works, so I now help other companies set up their own.

Context
A setup interview captures what you do, who you serve, your priorities and your history. After that every session starts already knowing the business.
Data
Your real numbers, pulled automatically every day instead of rebuilt by hand each month.
Intelligence
It watches your meetings, mail and numbers, then sends one brief with the part that needs you.
Automate
The recurring work eating your week gets audited, ranked, then handed over one task at a time.
Build
The time you win back goes into building what the business actually needs: your own integrations, internal tools, small apps, described in plain English. You stop being the operator and become the architect.

What it actually produces.

Illustrations of what the system puts in front of me every day, drawn from the real layouts with the contents left blank.

The morning brief

A brief lands at 04:00: what’s on today, what changed overnight, and where the numbers moved.

The command deck

One page for the whole business: revenue, capacity, pipeline, and what today needs from you.

Month-end invoicing

Invoices come out of the hours already logged, ready to check before anything is sent.

Two ways to get one.

The toolkit (for technical teams)

I package the system as something you install and run yourselves. Two companies already run it this way. If your team is comfortable with the tooling, this is honestly the better route: a system like this works best when the people using it own it end to end.

Done for you (for everyone else)

Interested but not technical? I handle the setup and build the system around how your business actually works. We find the workflows worth automating first, then grow from there. You own the result either way.

Start with one workflow, prove it, then expand. Problem first, AI second.

Ask your own data

An assistant that knows your numbers.

Your business already holds the answer: in a database, in the systems you run on, in a document nobody has time to read. Ask it in plain English and get it back in a sentence.

Knowing where the answer sits has never been the same as having it. The numbers might be in a database only two people can query, or on page 74 of something nobody will read. Meanwhile a general AI tool cannot help at all: it has no sight of your systems and no idea what your part numbers mean.

I connect the assistant to what you already have (a database directly, the systems you work in day to day, or your documents) and tune it to the language of your business. Your data stays where it is. A first working version reaches your team in weeks.

Try it on a sample business →Bring me your questions →

The demo runs on two made-up companies — a bottling plant and a wholesaler — so you can ask it anything. It answers five of the six questions above; the contract one reads documents rather than a database, which is a separate piece of work.

Currently building

Recent and active engagements.

Shopware

Ongoing

AI module for their Visual Management product, helping manufacturers find the biggest improvement opportunities in their production data. This is the assistant above, running in production: reading a live factory database and answering the shop-floor questions.

1Digit

Ongoing

AI engineer for a global mobility and cost-of-living data provider. Production AI pipelines covering 170+ countries, multilingual outreach in 11 languages, and the dashboards analysts use every cycle.

Who you’re working with

I’m Gerhard Bekker.

I started Simplifi AI because most businesses now sense that AI could change how they operate, but don’t know where to start or who to trust to actually build it.

My background is a bit unusual:

AI engineer at 1Digit, where I helped build production AI systems that gather and structure data on global mobility and cost-of-living, covering more than 170 countries.
PhD in Mechanical Engineering (Stellenbosch University). Five peer-reviewed papers on numerical modelling. I’m comfortable where the answer isn’t obvious.
Senior engineer at ENGYS, the company behind HELYX, the CFD simulation software used by multiple Formula 1 teams. I worked on the systems that made complex software reliable.
Data engineer at Spatialedge, where I built ETL pipelines for a financial services client at enterprise scale.

My work sits between technical rigor and business sense—the judgment to know which AI problems are worth solving, and which are not.

Based in George, South Africa. Open to clients anywhere.

LinkedIn ResearchGate

Curious whether AI can actually help your business?

Book a free 30-minute call. We’ll look at one workflow together. If there’s something worth building, I’ll tell you. If there isn’t, I’ll tell you that too.

Book a call →

Or reach out directly: gerhard@simplifiai.io · WhatsApp

Practical AI implementation,led by strategy.

Discovery call(free, 30 min)

Focused assessment(1 week, fixed fee)

Build and hand over(typically 2–4 weeks)

Context

Data

Intelligence

Automate

Build