Design
Interfaces people understand. Clarity is a feature, not decoration.
Design. Engineering. AI. Cybersecurity. Digital strategy. From the interface a person touches to the infrastructure nobody sees.
You need a digital system that makes the business faster, clearer and harder to ignore.
Interfaces people understand. Clarity is a feature, not decoration.
Systems that actually work. Data models, APIs, infrastructure, failure states.
Digital experiences designed around outcomes, not around screens.
Not screenshots. Each one is a problem, a system designed around it, and what the system does once it is running.
What happens after your phone leaves your hand?
Chronic illness is managed between appointments, not during them.
A till that refuses the sale is doing its job.
The connection drops. The shop does not close.
Entity-and-relationship investigation tooling. Link analysis over open sources.
Some systems are under client confidentiality and are shown privately.
I start with the problem. The interface is the last thing that gets decided, because it is the consequence of everything above it.
What is actually broken — stated in the operator’s words, not in features.
Where the work currently happens: the spreadsheet, the WhatsApp thread, the phone call.
The model. Entities, states, rules, and who is allowed to change what.
Data, services, integrations, failure behaviour and how it is deployed.
Only now. The surface that makes the system operable by a human under pressure.
Remove the repeated manual step. Every one removed is a class of error removed.
Instrument it. A system you cannot observe is a system you cannot improve.
Every digital object carries more than its content. A single file can describe the device that made it, where it was, when, and what it was edited with.
Individually trivial. Correlated, they describe a person.
My interest is in understanding what those signals mean — and building systems that protect people from what they reveal. Security is not a feature added at the end. It is a property of how the system was modelled in the first place.
Client work has to be finished. This doesn’t. The lab is where things are still being taken apart.
Where a language model genuinely replaces a manual step, and where it only appears to. Most of this work is deciding which of the two you are looking at.
Experiment write-ups — pending
Understanding exposure well enough to reduce it. What a system reveals in transit, at rest, and by accident — and what it would take to close that.
Experiment write-ups — pending
Most interesting systems are not one system. They are several, negotiating. Callbacks, idempotency, retries, and what happens when the other side is down.
Experiment write-ups — pending
Small pieces of software that delete recurring work. Reconciliation, notification, reporting — the tasks a person should not be doing twice.
Experiment write-ups — pending
Running the thing is part of building the thing. Servers, reverse proxies, process supervision, certificates, and the boring discipline of making deploys repeatable.
Experiment write-ups — pending
Code released publicly, and the reasoning behind it.
Things that do not fit a category yet. Usually the most useful ones later.
Experiment write-ups — pending
Grouped by what it is for, not by how long the list is. Everything here appears somewhere in the work above.
The product is what technology makes possible.
Technology should solve something real. If the problem cannot be stated plainly, the software will not fix it.
Complex systems should feel simple. The complexity belongs in the architecture, not on the screen.
Privacy and security are not decorations added at the end. They are decisions made at the beginning.
Before the frameworks. Before the APIs. Before any of it had a name.
There was a machine that wasn’t mine, an hour that was paid for by the minute, and the specific pressure of wanting to understand something faster than the clock allowed.
That constraint turned out to be the education. You learn to read a system quickly when you cannot afford to wander through it. Curiosity became design, design became development, and development kept leading one layer deeper — into the systems behind the screen, and then into the question of who else can reach them.
Based in Nairobi, Kenya. Building for African businesses and the infrastructure they actually run on.
Have a problem worth solving? Let’s turn it into a system.