We are a technology consultancy working at the intersection of software, data, and risk. We'll help you find the opportunity in the problem.
Three practices we run together on the same engagement, rather than handing off between teams.
Custom software built properly: applications, platforms, and integrations that run day after day, whether or not there is a model behind them.
The foundation everything else runs on: ingestion, storage, and governed data that analysts and applications can actually trust.
Quantitative models and the advisory around them, built to survive validation and regulator scrutiny.
Four stages, each ending in something you can hold us to.
We work out what decision you are actually trying to improve, what data exists behind it, and where the risk sits. Often this shortens the job.
Ends with a fixed-price proposalArchitecture and modelling approach on paper before any code. Definitions, assumptions, and what counts as success get agreed here, not argued about later.
Ends with a signed-off designShort cycles you can see. The pipeline, tests, and monitoring are built alongside the model rather than bolted on once it works.
Ends with a working system in stagingInto production, watched for drift, documented, and handed over. We stay on until it runs without us, which is the part most projects skip.
Ends with a system your team ownsRegulated, data-heavy sectors where the systems have to work and the numbers have to be right. Our home ground is South African financial services, and we work outward from there.
LocalMind is live in beta. A few more are close behind.
An AI platform that runs entirely on your own hardware. Documents are indexed, searched, and answered on site, so nothing leaves the building and nothing is sent to a third-party model provider. Built for organisations that have to answer to POPIA about where their data sits.
Pulls data from every system a business runs on into one governed source of truth, so every analyst and service works from the same numbers instead of multiple competing sources.
Enable automated data science capabilities. Autonomous data analysis, feature engineering and model building plus the documentation and validation trail needed to defend them internally and to a regulator.
Writing on systems, data, and risk, and the practicalities of getting the work into production.
The gap between a problem and a system that actually solves it is where value is made or lost.
CortechX is a small, senior technology consultancy in Johannesburg. We were set up to close that gap for organisations where the software has to work and the numbers have to be right.
Most firms either advise or build. We do both on the same engagement, which is what lets us take a problem from the first workshop through to a working system in production, and stay accountable for the result.
We keep the team deliberately small so the people who scope your work are the people who do it.
We are not a large generalist firm that assigns juniors once the partner has signed the deal. Every person on your project has done the work before.
Our clients come back because we deliver, not because we are good at writing proposals.
Tell us what you are working on. We reply within one business day with a straight view of whether we can help.
Thanks, we have it. You will hear from us within one business day. If it is urgent, call 063 501 9580.