Selected engineering outcomes
Selected initiatives across cloud modernisation, language technology, document intelligence, and applied machine learning—framed by the problem, the engineering decision, and the operating value delivered.
I led the redesign and migration of a VM-based deduplication solution to Azure Batch. Moving the workload to managed batch compute improved scalability and reduced annual client infrastructure costs by approximately $8,000.
Technical reference ↗
I owned the research and training of masked language models for Tshivenda and other indigenous Bantu languages. The work expanded language-technology coverage for languages underserved by mainstream NLP systems.
Technical reference ↗
I applied NLP and large language models to extract key information from regulatory documents. The workflow was designed to make review more structured, traceable, and less dependent on repetitive manual handling.
Technical reference ↗
I developed and tuned NLP models to identify and categorise job postings from social platforms. The pipeline converted unstructured posts into structured vacancy data suitable for downstream analysis.
Technical reference ↗
I developed multiple machine-learning models for predicting crop-disease outbreaks. The work connected model outputs to practical, data-informed pest-control recommendations.
Technical reference ↗
I designed and implemented an ETL and visualisation environment for public StackExchange data using Hadoop, Spark, and Tableau. The platform supported the complete analytical path from distributed processing to decision-ready dashboards.
Technical reference ↗