Skip to main content

Learning Center

Understand the systems, not just the outcomes.

Every lesson here is written to be checked, not just believed. Start with foundations, or jump to the track most relevant to your work.

Foundations

Beginner
12 min

What is a time series, really?

Stationarity, seasonality, and why price data breaks the assumptions most statistics courses start with.

Beginner
15 min

Separating signal from noise

A practical framework for deciding whether a pattern in market data is structure or coincidence.

Signal Processing

Intermediate
20 min

Feature engineering for market data

Turning raw ticks and bars into features a model can actually learn from, without leaking the future.

Intermediate
18 min

Denoising techniques compared

Wavelet filters, Kalman filters, and moving averages — what each assumes, and where each breaks.

Model Design

Advanced
25 min

Transformer architectures for time series

How attention-based models adapt to sequential, non-stationary data — and where they underperform simpler baselines.

Backtesting

Intermediate
16 min

Avoiding lookahead bias in backtests

The subtle ways future information leaks into a backtest, and the checks that catch it before you trust a result.

Advanced
19 min

Overfitting in financial models

Why a model that fits history perfectly is a warning sign, not a win — and how to measure real generalization.

Responsible AI

Beginner
10 min

Responsible research practices

How we frame assumptions, limitations, and uncertainty in every research note we publish.