Predictive Analysis of Internal Data

I’m looking for someone who can turn our raw internal database records into forward-looking insights. The data is already well-structured in SQL tables; what I need now is a predictive layer that helps us forecast key metrics and identify the drivers behind them.

You’ll have direct read-only access to the relevant tables along with a short data dictionary. From there, please choose the techniques and tools you’re most comfortable with—whether that’s Python (pandas, scikit-learn), R, or another established machine-learning framework—so long as the final model is reproducible and the code is clearly documented.

The deliverable I expect is a concise notebook or script that:
• cleans and preps the data,
• trains and validates at least one predictive model,
• outputs performance metrics and feature importance, and
• saves predictions to a table or CSV for downstream use.

When you respond, include examples of past work that demonstrate similar predictive projects; that will weigh heavily in my selection. I’ll be on hand to answer schema questions quickly, and I’d like to review a first cut of the model within a week of kickoff.