Valeriy Manokhin, PhD, MBA, CQF
Machine learning researcher, senior data science leader, and forecasting expert. PhD supervised by Professor Vladimir Vovk — the creator of conformal prediction.
Valeriy Manokhin, PhD, MBA, CQF, is a machine learning researcher, senior data science leader, and forecasting expert focused on probabilistic forecasting, conformal prediction, and production evaluation. He teaches modern forecasting to data scientists, engineers, and research teams, with students and readers from organisations including Amazon, Apple, Google, Meta, BlackRock, Morgan Stanley, and Nike.
His work on time series forecasting bridges academic benchmarks and the harder question practitioners face every day: which models still work when data shifts, constraints matter, and forecasts have to earn trust?
Education
- PhD, Machine Learning — Royal Holloway, University of London. Supervised by Professor Vladimir Vovk, creator of conformal prediction.
- MSc, Computational Statistics & Machine Learning — University College London (UCL).
- MBA — University of Warwick.
- CQF — Certificate in Quantitative Finance.
Books
Independent author and publisher. Most titles are sold direct on Gumroad — Mastering Modern Time Series Forecasting and Kiselev’s Arithmetic are also on Amazon.
- Mastering Modern Time Series Forecasting — this book. (also on Amazon)
- Applied Conformal Prediction — distribution-free uncertainty quantification.
- Mastering CatBoost — the gradient-boosting library, end to end.
- Foundation Models for Tabular Data — TabPFN, TabICL, and the new generation of tabular foundation models.
- Mastering Tabular ML: XGBoost Is Not Enough — GBDT vs. deep tabular vs. foundation models, benchmarked.
- Advanced Time Series Forecasting — companion to MMTSF, deeper into the modern stack.
- Forecasting Metrics That Don’t Lie — a focused treatment of the metrics chapter, expanded.
- Probabilistic Forecasting with Conformal Prediction — the intersection of forecasting and CP.
- Kiselev’s Arithmetic — the first complete English translation of Russia’s legendary arithmetic textbook. (also on Amazon)
The full catalogue — including bundles and Pro editions — is at valeman.gumroad.com.
Cohort courses on Maven
- Modern Forecasting Mastery — production-ready forecasting systems. 4.8/5 from 24 reviews.
- Applied Conformal Prediction — distribution-free uncertainty quantification. 4.7/5 from 37 reviews.
Open source
Maintainer of Awesome Conformal Prediction — the best repo on conformal prediction. Other repos at github.com/valeman.
Where Valeriy writes
- Substack: substack.com/@valeman — twice-weekly technical writing on forecasting and conformal prediction, paid and free tiers.
- Medium: medium.com/@valeman
- X: @predict_addict
- LinkedIn: linkedin.com/in/vmanokhin
- Bluesky: @predict-addict
- YouTube: @valeman_1
- Patreon: patreon.com/c/valeman
- Personal site: valeriy.ai
Research
- Google Scholar: scholar.google.com/citations?user=e_AaIGYAAAAJ
- ResearchGate: researchgate.net/profile/Valery-Manokhin
About this book
Mastering Modern Time Series Forecasting exists because most forecasting writing is either too academic to ship or too superficial to trust. The book is built around a simple principle — forecasting models are only 5% of the equation. The other 95% (metrics, validation, deployment, failure modes, real-world constraints) is what separates a working forecast from a hallucination.
The book is a living resource. Many improvements and additions come directly from reader feedback, and contributors are credited in the acknowledgments.
For book questions, errata, course enquiries, and speaking: modernforecastingmastery@gmail.com