Machine learning engineer with systems depth across software, data, and embedded work.

The largest current build is Fyndel, an AI-powered construction finance platform with WIP reporting, secure multi-tenancy, workbook intelligence, predictive forecasting, and accounting integrations.

It has been a useful systems project because the data model, security rules, calculations, imports, dashboards, ML pipelines, and inference services all need to stay consistent across tenant-scoped financial data.

The LORIS satellite project demanded careful C development, communication protocols, test discipline, and clear documentation under real system constraints.

That background carries into ML work as stronger interface boundaries, better failure awareness, and a preference for systems that remain understandable as complexity grows.

NLP for training-needs analysis, anomaly detection for aircraft health monitoring, and intrusion detection on large network datasets are the clearest completed ML examples.

I care as much about the surrounding system as the model itself: the pipeline, evaluation loop, API behavior, and delivery surface that determine whether the work is actually usable.