Training
Data Analysis with Python & SQL
From raw exports to decisions: SQL that answers real questions, pandas without the flailing, visualization that communicates, and the statistical literacy to know when your numbers are lying.
Every company is sitting on data it doesn't use, and the gap is rarely tooling — it's people who can move from "here's a CSV" to "here's what we should do". This program builds that skill honestly: SQL as the first-class analysis language it is, Python/pandas for what SQL can't reach, visualization as communication rather than decoration, and enough statistics to avoid the classic self-deceptions (survivorship, confounding, torturing averages).
Everything is exercise-driven on realistic messy datasets — because real data is messy, and cleaning it well is half the profession.
Curriculum
-
1
SQL for analysis: joins, aggregations, window functions, CTEs
-
2
Python and pandas: loading, cleaning, transforming without copy-paste chaos
-
3
The dirty-data survival kit: nulls, duplicates, encodings, dates
-
4
Exploratory analysis: distributions, cohorts, funnels, retention
-
5
Statistics for the honest: significance, confidence, common traps
-
6
Visualization that communicates: chart choice, labeling, narrative
-
7
Capstone: a full analysis from raw export to executive summary
