Johnathon H.Private profile: Upwork shows its earnings only to signed-in members, so we check it every few weeks, not daily. MRR here is a rough estimate (Upwork's six-month earnings ÷ months active). Last read Nov 5, 2026.

I turn messy data into models you can trust. I'm a Kaggle Expert ranked in the top 0.3% of 217,000+ ranked competitors, with 1 gold, 1 silver and 6 bronze medals across competitions and notebooks, and a top-10 UK notebook ranking. I also hold an MSc in astrophysics, did PhD-level research in Astrophysics with Bayesian inference, and have qualified as a Google Professional Machine Learning Engineer. I get up to speed on new datasets fast, and I'm happiest with unusual data: sensors, time series, scientific measurements, imaging, survival data. What I deliver - Predictive modelling: forecasting, regression and classification on tabular and time-series data, from baseline to tuned, validated ensembles. - Sensor and IoT analytics: normal-behaviour modelling, fault and anomaly detection, condition monitoring. - Causal and Bayesian analysis: answering "did the change work?" with credible intervals instead of before/after guesses. - Production-ready delivery: reproducible pipelines, experiment tracking, ETL automation and deployment on Google Cloud. Results - Automated fault-severity classification, saving a team 20%+ of monthly reporting time. - Bayesian causal uplift analysis that informed upgrade rollout decisions across a wind-turbine fleet. - Sensor models predicting gear-oil temperature to within fractions of a degree. How I work: a short call, then a data review and baseline in the first few days. We agree a plan tied to the metric you care about, and I share progress and a written summary as I go. I also build LLM/RAG and agent systems on Vertex AI when a project needs them. Stack: Python, XGBoost/LightGBM/CatBoost, PyTorch/TensorFlow/JAX, scikit-learn, PyMC, CausalPy, Optuna, MLflow, W&B, GCP.