Writing
Field notes
Short answers to the questions we are asked most about preparing financial data for models, each with a test you can run without us.
Embeddings Choosing an embedding model for financial documents, and what goes wrong when you do not
Six failures, from a model that thinks delta is an airline to a vendor that retires the model your whole index depends on.
Read notePoint-in-time Why your backtest is optimistic
Point-in-time data explained with one restated AUM figure, and how to check your own history for the same problem in an afternoon.
Read noteRetrieval Why an embedding search cannot find a CUSIP
Identifiers are strings without meaning. What a vector index does with them, and the two-line fix.
Read note