Initializing Redox Engine
Reduction ◆ Oxidation
Software systems engineered at the reaction point. We build where opposing forces meet — frontend and backend, speed and stability, design energy and engineering depth.
Client work
Commissioned builds and advisory work. Fewer engagements than a big agency lists, each one we can talk through line by line.
One marketplace, four audiences, built end to end
Customers ordering, stores selling, couriers delivering and admins running the whole thing — each with their own surface, over a single API. Built from an empty repository to a system that handles catalogue, orders, payments, earnings and payouts.
Studio products
We are a product studio that takes client work. These are ours — shipped, maintained, and the reason we know what we are selling.
Anonymous messaging and tipping for creators
Cilbup — "public" reversed — lets anyone send a creator an anonymous message, and attach a tip to get priority attention. Fans pay with crypto or fiat; creators earn in USD, on a platform that installs to a phone without an app store.
School management where the school keeps its own money
Academics, staff, students, timetabling and fees for African institutions, split across four persona surfaces so no screen serves two audiences at once. Fees are collected on each school's own payment credentials — the platform records the money without ever holding it.
An AI layer over the task manager you already use
Resurgee augments Google Tasks and Calendar rather than replacing them. It watches for tasks about to slip, suggests rescheduling options when they do, and plans your day through a conversational assistant — across multiple linked Google accounts.
Capabilities
We do not sell isolated deliverables. Every capability below exists because the others need it.
The process
Four phases, each with a written artefact you keep whether or not we continue.
We start by understanding where you are and what you are trying to reach.
Whether you have a live product, a prototype, or just an idea written down, the first work is the same: we map the goal, the users, the constraints, and the current state — through interviews and a read of whatever already exists. You leave with a written analysis and a recommended path, whether or not we build it together.
The plan, written down — with the trade-off behind every decision.
System design, data model, scope, and milestones as reviewable documents. You approve the shape and the sequence before engineering starts, and the decision records travel with the codebase so nothing is built from an assumption nobody wrote down.
Two-week iterations, with a working preview from the first week.
You get a live preview environment early and a demo every two weeks. Tests, CI, and monitoring are part of the definition of done, not a hardening phase promised for later. We release to production when a milestone is genuinely ready — the two weeks are a rhythm for feedback, not a countdown to launch.
We hand over a system your team can run without us.
Runbooks, architecture walkthroughs, and paired on-call until your engineers are comfortable operating it. The engagement ends when the system is stable in your hands — that is the equilibrium we are aiming for.
The blog
What we found in the code, and the reasoning behind the calls we made.
Start here
Tell us what you are building and what is currently in the way. We reply to every enquiry within two working days — with a real answer, not a calendar link.
