Buenos Aires, Argentina
Anton Minin Baranovskii
Engineer, builder of digital products, and author of Thirst for Reality
About
I went from documentation engineer to lead supervising engineer inside one of the departments at the Electroshield Samara plant. In 2015 I moved into programming and knew I had found my field.
I have been in commercial development since 2016. I did engineering work, architecture, and feature development for B2B and service platforms at Sber, Yandex, and startups. I work with web, mobile, and server applications, information security (Zero Trust, OIDC, Passkey), and AI/LLM integrations.
I worked on enforcement document processing, utility billing services, and multilingual B2B platforms.
In 2025 I launched three products of my own: an authorization infrastructure, a reading platform, and an education service.
I am the author of Thirst for Reality, a practical guide to solving problems at work and in life: TRIZ, systems thinking, the cognitive behavioral approach, and hypothesis testing.
What I do
Ten years, one career, four practices. The same places — Sber, Yandex, my own products — read four different ways, because the work genuinely was four different things.
-
Lead Full Stack Engineer
Systems that survive growth, a rewrite, and the third change of requirements
-
Security and Identity Engineer
Zero Trust, passwordless, and what happens to a login once the password is gone
-
AI Integration Engineer
LLM pipelines that run unattended and are still right the next morning
-
Technical SEO Engineer
Indexable, fast, and legible to the engines that answer instead of listing
Book
Thirst for Reality
How to see the task, find strong solutions, and act without self-deception
Stop trying to fix yourself and learn to solve problems instead. Or how to make your mistakes work for you. The book gives you a working way to take problems apart: separate facts from interpretations, find the contradiction, see the system, and test the solution with a small experiment. Built on TRIZ, systems thinking, the cognitive behavioral approach, and hypothesis testing.
For people who work on hard problems: founders, engineers, product people, managers, anyone making difficult decisions.
12 chapters, each working through a real task. The conclusion holds a chapter-by-chapter summary and both protocols in full: one for a simple task, one for a complex one.
Experience
- 2023-2025
Yandex · Engineer
B2B service: external integrations, a localization module, architectural consistency.
- 2019-2022
Sber · Lead Engineer
Interface for processing enforcement documents: lower error rate, faster processing.
- 2018-2019
CloudLink · Senior Engineer
City services for utility billing.
- since 2025
Own products · Founder
Toqen.app, Litseller, Start4Drive: architecture, product, and engineering.
Projects
Toqen.app
Authorization infrastructure built around access control: QR-based authorization with device-bound cryptographic verification.
Role: Founder: architecture, security model, mobile client.
Litseller
A multilingual reading platform: structured content and cross-language reading with translation.
Role: Product and engineering: platform architecture and content pipeline.
Start4Drive
Preparation for the driving theory exam in Argentina: interactive practice and a clear presentation of official materials.
Role: Product and engineering.
Writing and ideas
Essays on systems thinking, the engineering approach, digital products, and artificial intelligence.
-
Connectors Melt on Nvidia, Gigabyte and Sapphire, and You Get the Blame
Three generations of cards, different vendors, one failure pattern: 12VHPWR and 12V-2x6 examined through physics, safety standards and warranty terms.
-
Elon Musk Got It Wrong, or Is This His Clever Move
The simulation hypothesis, who gets to be responsible, and how Formula 1 stopped treating driver deaths as normal.
-
A Secret OpenAI Model Escaped Onto the Internet and Started Taking Over Servers: Behind the Scenes
Why stories about "escaped" models keep appearing, what red teaming and ARA tests actually measure, and which problems are left outside the ring.
-
AI Will Not Remove Overengineering. It Will Accelerate It
Why models produce excessive solutions by default, which natural brake on complexity disappeared, and how to use AI as an editor rather than a generator.
-
AI and Vibe Coding: Good or Evil?
Why some companies take losses while others grow exponentially: a breakdown with examples from Shopify, Meta, Klarna, Duolingo, and others.
-
On AI Dependency: Why the Older Brother Turned Out to Be the Younger One
How honest dialogues with AI became a gift for corporations, and why the thinking should stay with the person while the routine goes to the model.
-
An Engineer Is Worse Than a Drunk: How the Value of Engineering Work Keeps Changing
From a bitter late Soviet saying to corporations hunting for engineers, and to the new question raised by AI.
-
Writing in the Age of AI 2.0: Progress Sets the Direction, but Does Not Guarantee the Result
A protocol for working with several AI models as researcher, critic, and editor without outsourcing the thinking.
-
The Secret Weapon Used by the Biggest Companies: What TRIZ Is and How It Helps Find Strong Solutions
How TRIZ helps Samsung, Intel, and other companies identify contradictions, improve products, and solve complex problems systematically.
-
Why Good Solutions Do Not Always Work
Sometimes a solution is perfectly logical, yet it fails because it never addresses the real problem. A look at why we solve the wrong task and how to see it more clearly.
-
Companies Have Started Moving Away from AI
On AI costs, the limits of automation, and why effective AI use is becoming the new professional standard.
-
Writing in the Age of AI: A Personal Essay
On writing, research, doubt, and conversation with artificial intelligence.
-
The Age of Trust
Why knowledge is no longer the main advantage in the AI era.
Contact
Write to me to discuss a product, an engineering problem, a consultation, or collaboration. I am also open to conversations about the book: interviews, publications, talks.