I design and ship AI systems for environments where the system has to be governable before it is allowed to be useful.
Assistants and agent systems designed for oversight — human checkpoints, audit trails, and clear failure behaviour built in from the start.
Pipelines, warehouses and batch systems that hold up under regulatory scrutiny and still make sense to whoever inherits them.
Azure-first design and cost discipline, with multi-cloud where the constraints genuinely call for it.
| Position | Focus | Period |
|---|---|---|
| Founder — Rooral Platforms | Data and AI infrastructure for Southern Africa | 2025 — |
| Enterprise AI — Financial services | Applied AI and data science in a regulated group function | 2022 — |
| Software engineering — Consulting | Public sector modernisation; distributed batch systems | 2018 — 2022 |
| MSc Big Data Science | University of Pretoria | 2023 |
| BScEng Electrical | University of Cape Town | 2016 |
One considered piece a month on applied AI, cloud and data engineering. Written for practitioners.