Software engineer with a Master's in Signal Processing and Machine Intelligence. I specialize in applying AI and ML directly into production software — designing systems that learn, adapt, and perform at scale. Always connecting and working globally.
We're entering an era where AI is not just a feature you ship — it's the engineering primitive you build with. From code generation to intelligent pipelines, here's what that shift really means for engineers on the ground.
Fourier transforms, convolutions, and filtering — these aren't just DSP tools. They're the hidden mathematics powering modern deep learning.
↗Training a model is 20% of the work. The other 80% is making it survive contact with real users, real data, and real infrastructure.
↗Transformers aren't just for text. Here's how attention mechanisms can be applied to sequential signal data with remarkable results.
↗Raw signals are rarely what your model needs. A hands-on breakdown of spectrograms, MFCCs, wavelets, and why they matter for ML.
↗After integrating LLM-powered code review into a production pipeline, here's what actually worked, what failed, and what surprised me.
↗Microservices, model serving, feature stores, and event streaming — how to architect ML systems that don't collapse under their own weight.
↗Why great research papers become terrible software, and how engineers with academic backgrounds can bridge that divide intentionally.
↗Building low-latency audio intelligence systems means making hard tradeoffs between accuracy, speed, and memory. Here's my framework.
↗The global AI race needs builders everywhere. Why the broader African tech ecosystem are positioned to lead — not just participate.
↗