SQL & Data Wrangling
Joins, aggregations, and window functions pull clean features out of the warehouses where market data actually lives.
Most real market data lives behind a SQL interface in a warehouse — not in files on your laptop. This node covers the query patterns you will use daily: filtering and aggregating with GROUP BY, combining tables with joins, and the window functions — row_number, lag, moving averages over partitions — that turn raw ticks into aligned features. You also learn the wrangling mindset: profiling a dataset for gaps and bad ticks, deduplicating corporate actions, and deciding what to pull versus what to compute downstream in pandas or Polars. Solid SQL multiplies the value of everything else in this track, because every dataset starts as a query.