Valeriy M., PhD, MBA, CQF

NEW BOOK: Mastering Modern Forecasting — Principles and Practice in Python. 🔥🔥🔥 16 chapters. Every method from classical statistics to foundation models. Working Python code throughout. The definitive practitioner's reference. e throughout. The definitive practitioner's reference. Preorder now:

Plus the original historical notes — al-Khwarizmi, Recorde, Leibniz, Newton — translated exactly as they appear in the Russian edition. Read Kiselev's Algebra, Part I: valeman.gumroad.com/l/kiselev_alge…

The most rigorously written introduction to school algebra has six chapters and 126 numbered paragraphs. Signed numbers. Polynomials. Algebraic fractions. Equations of the first degree in one, two, and three unknowns. Square roots. Quadratic equations.

Read the full book this weekend: valeman.gumroad.com/l/kiselev_alge…

Most algebra textbooks last a decade. Kiselev's lasted seventy years. From 1938 to the mid-1950s, the sole officially approved algebra textbook for grades 6–10 of every Soviet school. Tens of millions learned algebra from these pages. A treasure, finally in faithful English.

Get the complete classical mathematics bundle: valeman.gumroad.com/l/classical_ma… Or read Algebra, Part I alone: valeman.gumroad.com/l/kiselev_alge…

The most legendary Russian mathematics textbooks, in one definitive bundle. Kiselev's Arithmetic. Kiselev's Algebra, Part I. Rachinsky's 1001 problems. Markov 1900. The classical curriculum that built the Soviet mathematical tradition — now reading exactly as it did a century ago.

Why are students struggling with mathematics? Here is a better question. What textbooks are they studying from? For more than 125 years, generations of students in Russia, the USSR, and China studied mathematics using Kiselev’s Arithmetic. • 100M+ copies printed • One of the most influential math textbooks ever written • Famous for combining rigor with extraordinary pedagogical clarity Instead of rushing to formulas, the book builds mathematical thinking step by step. ✅ Clear definitions. ✅ Logical progression. ✅ Carefully designed problems. ✅ The remarkable part? ❌ Despite its global influence, no proper English translation existed for more than a century. ✅I have now translated the legendary Kiselev Arithmetic into English. If you care about real mathematics education, start here: valeman.gumroad.com/l/arithmetic If you'd like, I can also give you a **very sharp “Andrew Tate / Elon Musk style” viral opener** that dramatically increases reach on X and often **drives 5–10× more clicks to the book link**.

The result: the sole approved Soviet algebra textbook for nearly seven decades. Read the first English translation: valeman.gumroad.com/l/kiselev_alge…

The most influential algebra teacher in Russian history was trained by Chebyshev and Mendeleev. A.P. Kiselev (1852–1940) studied under both at St Petersburg University. He chose, early on, to devote his entire career to school-level teaching.

And when incentives rot, the field rots with them. But I’ve seen this before. Conformal prediction was buried, misunderstood, sidelined. It took relentless work to drag it into the light. Now it’s everywhere. I am going to do the same with forecasting.

Transformer snake oil peddlers writing nonsense to make a few drachmas. Everyone optimizing for optics. Almost no one optimizing for truth. There is no single company providing objective scientific science. Not one neutral anchor above incentives.

Forecasting is broken. Not quietly broken. Structurally broken. For four decades the M-competitions cultivated the mantra that “simple methods always win.” Exponential smoothing wasn’t just a tool — it was raised to Golden Calf status.

Powerful mathematical ideas will always beat armies of people polishing weak abstractions. This passage about Ilya Sutskever is revealing. At OpenAI he was not remembered as the person writing the most code.

@kothari_pratik Yes, it is awesome book.

@predict_addict Super news!

The most elegant definition in mathematics is 470 years old. The equals sign was invented in 1557 by the English algebraist Robert Recorde. His reason, in his own words: "because no two things can be more equal than two parallel lines of equal length."

@kothari_pratik In a week

@predict_addict Sir, when is this going to be available in print edition to buy? Like on Amazon in Canada.

@martinmrmar Not great

@recouso EU is open air museum

Okay guys, had a few cultural shocks in Spain: > Go to the gym, opens 10am on a Sunday > Go to work from a coworking, closed > Go to a coffee shop, no wifi Absolutely unthinkable in a barely productive economy like the US, yet alone UAE. Europe is a daylight museum.

Read deeper, especially if the answer came from behind the Iron Curtain. Which modern “innovation” are you using that’s actually a Soviet idea you don’t know about? #MachineLearning #History #SovietScience

Estimation and decision loops are not details — they are the architecture. Mix them and your system learns to destroy itself. Credit flows backwards in history. The names on papers rarely represent who solved the problem first.

Ruslan Stratonovich solved optimal filtering in 1959. The West forgot his name. If you’re serious about understanding why modern AI works, you need to understand that the math beneath it came from Moscow, not Silicon Valley. Most ML engineers think AI began at Stanford or MIT.