Latest Thinking

All posts
tradingtrading

Part 1Time-Series Foundations: Transformers, Diffusion, and Why They're Different

Before we build anything, we need a clear mental model of what these architectures actually do — not the marketing version, the mechanistic one.

Jun 2, 2026Read →
tradingtrading

Part 2Data Pipeline: NASDAQ-100, Options Chain, and Sentiment

How to align three different data sources into one consistent daily dataset — and why every alignment decision is also a leakage decision.

Jun 2, 2026Read →
tradingtrading

Part 3Feature Engineering: From OHLCV to a Rich Feature Matrix

What each feature actually measures, why it might matter for predicting NASDAQ-100 returns, and how to compute it without accidentally injecting future information.

Jun 2, 2026Read →
tradingtrading

Part 4The Standard Transformer: Architecture, Training, and Why It Collapsed

We built it. We ran it. It predicted the same thing for every stock. Here's exactly what happened and what it reveals about applying transformers to financial data.

Jun 2, 2026Read →
tradingtrading

Part 5Temporal Fusion Transformer: Gating, Memory, and Sector Learnability

The architecture that finally worked — and a precise explanation of why each component addresses a specific failure mode of the Standard Transformer on financial data.

Jun 2, 2026Read →
tradingtrading

Part 6Validation Discipline: Catching Leakage, Rank IC, and the Economic Viability Wall

The step most AI trading projects skip — and the reason most of them fail when they reach real capital. A rigorous validation framework for quantitative ML.

Jun 2, 2026Read →
tradingtrading

Part 7LightGBM as Signal Engine: Feature Sets, Clean Candidates, and OOS Testing

The model that finally survived adversarial validation — and why gradient boosting often beats deep learning on structured financial data.

Jun 2, 2026Read →
tradingtrading

Part 8System Architecture: Signal → Strategy → Execution

A statistically real signal is not yet a tradeable system. This is the architecture that converts +0.08 Rank IC into a deployable paper-trading framework — and why the layers between prediction and profit are harder than the model itself.

Jun 2, 2026Read →

Browse Topics