HomeData ScienceAI Studio

lab

Smaller work: notebooks, Kaggle attempts, visualization exercises, and learning experiments. A lab notebook, not a highlight reel, including what didn't work.

Filling up as I go

These entries describe planned experiments and will be updated with links and lessons as each one is completed.

  • Tableau Public replica of the retail dashboard

    Aug 2026Planned
    Explored:
    Rebuilding the four-view retail dashboard in Tableau Public from the same star-schema model, to add a named BI tool alongside the Streamlit version.
    Learned:
    Planned, the data model and KPI definitions are already documented to transfer.
    • Tableau
    • SQL
  • Porting the SQL analytics project to PostgreSQL

    Aug 2026Planned
    Explored:
    Translating the SQLite warehouse and window-function queries to PostgreSQL (DATE_TRUNC, QUALIFY) to practise warehouse-dialect SQL for interviews.
    Learned:
    Planned.
    • PostgreSQL
    • SQL
  • Uplift extension to the churn model

    Aug 2026Planned
    Explored:
    Moving from predicting who is likely to churn to who is persuadable, the honest next step for the retention use case, once outcome data is available.
    Learned:
    Planned.
    • Python
    • scikit-learn