I will build machine learning models and statistical analysis in python

Data Scientist with production machine learning experience predictive modeling, statistical analysis, and survival/time-to-event methods, built for real decision-making, not just notebook demos.

Background: built NLME and survival models for clinical research that reduced required sample sizes by 30-40%, predictive models for cardiovascular risk outperforming standard clinical scales, and statistical pipelines supporting drug efficacy comparisons across markets. 2 peer-reviewed publications in data analysis.

What I can help with: building and evaluating ML models (Random Forest, XGBoost, regression) with proper validation, not just a fitted model with no diagnostics; survival/time-to-event analysis (Cox PH, Kaplan-Meier); A/B testing and statistical significance testing done correctly; data cleaning and feature engineering; explaining methodology clearly enough that you can defend it yourself in a thesis defense or stakeholder meeting.

Stack: Python (Scikit-learn, Pandas, NumPy, SciPy, PyTorch), R, Statsmodels. Every deliverable comes with a plain-language explanation of what was done and why. Send your dataset or describe it, and I'll suggest the right approach