Data solutions, engineered like real software

I design and build data platforms, pipelines, dashboards and machine-learning solutions for South African businesses. The work ships as production software your team can run without me, never as a throwaway notebook.

I’ve worked with amazing companies

Sound familiar? - Recognise any of these?

Most engagements start with a symptom rather than a strategy. These are the six I see most often in South African businesses. If one of them sounds like yours, ten minutes of free scoping puts a ZAR cost band on it, no email required.

Manual data work eating hours every week. Someone copies, reconciles or chases the same numbers every week, and the hours have stopped feeling fixable.
Spreadsheets that have outgrown themselves. What began as one tab is now a fragile workbook only one person understands.
Reports that take days to assemble. By the time the monthly report is finished, the month is already half over.
Dashboards your team checks but doesn't trust. The chart says one thing, the gut says another, and the team quietly stops looking.
A legacy system nobody understands anymore. It mostly works. The person who built it has left. Touching it is scary, and so is leaving it alone.
Data in five places that reconciles in none. CRM, finance, ops and one Google Sheet, each with its own version of the same customer.

Put a ZAR number on it: free scoping

Selected work

Real examples of my work, from a continental solar-fleet data platform to offline-first invoicing, for businesses in energy, financial services and property.

2026Case study

From Inverter to Impact: Engineering a Continental Solar-Data Platform

A production data platform that unifies 15 inverter ecosystems, processes global solar irradiance in C++, and surfaces fleet carbon impact and O&M health across 24,350 solar plants.

2026Case study

What Drives the Price of Food? Oil, Fertiliser, El Nino and the 2026/27 Squeeze

An interactive analytical study over decades of public data: how oil, fertiliser and El Nino move South African food prices, what Washington does and does not explain, and a dated forecast for 2026/27.

2026Case study

One Core, Three Shells: Invoicing Engineered Like Real Software

An offline-first, SARS-compliant invoicing and bookkeeping product for South African small businesses: one Zig core linked into desktop, web and native shells, with integer-cents money and 334 automated tests.

Few people show such passion for understanding their client and their business, but Ridhwaan does. I had the pleasure of working closely with Ridhwaan for almost a year. Ridhwaan was hands-on as he managed the project and provided valuable analysis. I was particularly impressed by Ridhwaan's ability to unpack data into meaningful insights for the business. I look forward to working with Ridhwaan in the future. He provides true value to any client. - Ahmed Cassim, MD at Hello Group

Services - Three services covering the full data value chain

Everything I take on falls under one of three services. Each one ships as production software that your team can run, audit and extend after I leave.

Data Engineering. The foundation layer. I build pipelines that pull data out of your systems and vendor APIs, a warehouse that holds a single version of the truth, and the data models and capture forms that keep it clean. For Decarb.Earth this meant unifying fifteen inverter ecosystems into one pipeline covering 24,350 solar plants.
Data Analytics. The decision layer: dashboards, data visualisation and business intelligence. I set up open-source BI tools such as Metabase on fixed-cost hosting, or Power BI where your business already lives in Microsoft 365, and I handle the deeper diagnostic analysis when the question is why a number moved rather than what it is.
Data Science. The predictive layer. Forecasting, clustering, pricing models and machine learning, deployed as monitored services on top of the first two layers. Recent examples include a hierarchical pricing model for short-term rentals and multilingual sentiment analysis for an insurer.

From the blog

Practical writing on data engineering, analytics and building data solutions like real software.

API Ingestion at Scale: Lessons from Unifying 15 Inverter Vendor APIs

July 3, 2026

Practical lessons from building a Go ETL pipeline that ingests solar telemetry from 15 inverter vendor APIs across 24,350 plants: rate limits, data quality, idempotency.

Do You Need a Data Lake? DuckDB and MotherDuck as the Pragmatic Alternative

June 26, 2026

Most businesses told they need a data lake or lakehouse don't. When Parquet + DuckDB or MotherDuck covers your analytics, and when Postgres alone is enough.

Tell me about your business challenge

Contact Ridhwaan

My offices

Johannesburg
South Africa
2198