About Stefan Jansen

ML4T is written by someone who builds these systems for a living — in markets, and well beyond them.

Stefan Jansen

Stefan Jansen

Author, Applied AI founder, and ML practitioner

Stefan Jansen is the author of Machine Learning for Trading and the founder of Applied AI. The book grew out of investment practice. Since 2013, his work has included building an investment data-science office at Infusive Intelligence, designing quantamental research workflows and event forecasts for a public-equities strategy, serving on an investment committee, and later building research and live-trading systems through Applied AI. That experience shapes the book's focus on the full path from data and labels to validation, backtests, portfolio construction, costs, and live operation.

Background

  • Author, Machine Learning for Trading (1st, 2nd, and 3rd editions)
  • Founder of Applied AI; production ML systems since 2016
  • Maintainer of the ML4T open-source ecosystem and companion libraries
  • Harvard master's in economics and public policy; Georgia Tech MS in computer science; CFA charter

The third edition is more than a book update

Nine cross-asset case studies, companion libraries, primers, diagnostics, and courses make the workflow something readers can run and inspect.

1st
2018

Foundations of ML for trading

2nd
2020

Deep learning and NLP for finance

3rd
Latest
2026

Cross-asset case studies and production workflow

Written from the working process

A model forecast is only one step in a trading workflow. The harder questions are usually around data timing, label design, validation, costs, portfolio construction, monitoring, and the decision to stop, retrain, or deploy.

That is the practical center of ML4T. The third edition follows the path from research idea to evidence and then to live operation, with case studies across ETFs, crypto perpetuals, microstructure, options, futures, FX, firm characteristics, and US equities.

Stefan's day work reaches beyond trading — into contract intelligence for insurers, forecasting in healthcare, and AI agents through Applied AI. That range shapes the book's point of view: a model is only as good as the data, evaluation, and operating process around it.

19,000+
GitHub Stars
27
Chapters
446
Notebooks
3
Editions

Working on implementation?

Applied AI is Stefan's consulting practice for teams building production AI and machine-learning systems: strategy, architecture, code, evaluation, and operation.