As an Associate QA Engineer at Numerator, I own the accuracy of consumer panel data that powers market research decisions for some of the world's largest brands. Bad data doesn't just slow down pipelines — it breaks trust. I've built the processes, validations, and root cause frameworks that stop bad data before it reaches stakeholders. On the science side, I build. Regression models. Classification pipelines. Time series forecasts. End-to-end ML workflows using Python, Pandas, and scikit-learn — with evaluation metrics that actually mean something (RMSE, AUC, accuracy) because a model no one can measure is a model no one should deploy. I'm also actively developing skills in prompt engineering and LLM interaction design — bridging structured data thinking with generative AI. What I bring to your team: → Data validation & quality frameworks (SQL, Excel, Python) → Predictive modelling: churn, classification, regression, forecasting → EDA, feature engineering, and preprocessing pipelines → KPI monitoring and market research reporting → Clear communication of data findings to non-technical stakeholders I don't just find problems in data. I build systems that prevent them.
Model Development & DeploymentClassificationRegression+31
