Book Cube: 2026, page 5
Book Cube publications from 2026, page 5: bilingual notes, sources, and original Telegram posts.
- · #BookCube
In the afternoon IT Picnic, and in the evening the match CSKA - Rostov. Very busy day and wants to sleep, but once promised, I had to go to football.)
In the afternoon IT Picnic, and in the evening the match CSKA - Rostov. Very busy day and wants to sleep, but once promised, I had to go to football.)
- · #AI4SDLC
Materials on the second AMA session about AI-assisted Engineering with Alexei Litvinov about how to implement AI in development (Category AI4SDLC)
Materials on the second AMA session about AI-assisted Engineering with Alexei Litvinov about how to implement AI in development (Rubric AI4SDLC ) Ready materials from the live podcast Code of Leadership with Alexei (@tip\ podcast):
- · #AI4SDLC
AI4SDLC 2026How AI is changing software development in Russia (Category AI4SDLC)
AI4SDLC 2026How AI is changing software development in Russia (Rubric AI4SDLC ) Today at IT Picnic I will talk about our AI4SDLC research. 2026 and the general state of AI in software development. We’ve already gone through a phase with assistants, and it’s exciting to see if agent-based development helps not just speed up coding, but also lead to faster and more sustainable product delivery. Or are agents simply shifting bottlenecks to task setting, review, testing, reworking, and operation?
- · #Learning
Understanding is a new bottleneck in working with AI (Category Learning)
Understanding is a new bottleneck in working with AI (Rubric Learning ) I saw it. 20minute speech Geoffrey Litt, design engineer at Notion. AI accelerates the production of change, but not the formation of a system model in the head. So the bottleneck is no longer writing code, but human understanding. And it is important not only to check the results. Agents also learn to test and review themselves. Understanding is necessary to participate: noticing connections, proposing the next move, and being responsible for the development of the system.
- · #AI4SDLC
In five minutes, the live edition of “Research Insights Made Simple” about AI development as an evolving stack will begin.
It starts in five minutes. live “Research Insights Made Simple” is about AI development as an evolving stack. Why is the same model in two coding agents so different? And what should a company really consider its AI stack: model, bandage, tools, data, or the right of an agent to change internal systems?
- · #AI
There was a recording of my performance at Turbo ML Conf.
Appeared. record I'm here at Turbo ML Conf. The slides themselves are available on My website is polomodov.tech.. P.S. Materials on all tracks are all published and ready to view:) 1️⃣ Fundamental Advances & Exploratory R&D - Vkontakte YouTube 2️⃣ Applied ML at Scale & Business Impact - Vkontakte YouTube 3️⃣ ML Infrastructure, Platforms & Engineering Core - Vkontakte YouTube
- · #AI
Jensen Huang: From Three Textbooks to a Universal Approximator (AI column)
Jensen Huang: From Three Textbooks to a Universal Approximator (Rubric AI ) I saw it. conversation Harry Tana with Jensen Huang at Startup School 2026published 26 July 2026 years. Usually, such interviews quickly turn into a set of success stories, but here one engineering habit passes through the whole conversation: to recognize that the old model does not work, to understand a new area and see the change of the entire computing system behind a particular technology. 1The first story is about the mistake that started NVIDIA.
- · #AI4SDLC
SonarQube Cloud: Why Agent Code Needs Deterministic External Control (Category AI4SDLC)
SonarQube Cloud: Why Agent Code Needs Deterministic External Control (Rubric AI4SDLC ) After the post.Sonar and the Star Hour of VerifiersI decided to find out what exactly. SonarQube Cloud Checks in the code written by the agents, and also why it is not another printer, but something more, for example, a gate between “the agent is finished” and “the change can be poured”.
- · #AI4SDLC
Join our livestream with Ivan Gel on UpCore, a digital team lead
Questions for Ivan Gel about UpCore: effectiveness, seniority, AI-assisted coding, the evidence behind its claims, team trust, and the boundary between telemetry and surveillance.
- · #Books
Tanenbaum, PagedAttention and the Foundation That Doesn't Get Old (Books column)
Tanenbaum, PagedAttention and the Foundation That Doesn't Get Old (Rubric Books ) I was dealing with an interesting paper.Efficient Memory Management for Large Language Model Serving with PagedAttention" from the creators of vLLM. In 2023 In the year, the authors looked at how modern LLM inferencing engines work with memory, and found an almost criminal picture for the present time. (The efficiency killer was KV-cache.). The systems pre-reserved a continuous memory portion based on the maximum possible sequence length.
- · #AI
Launching a podcast 3 AImigo: three views on AI and development (AI column)
Launching a podcast 3 AImigo: three views on AI and development (Rubric AI ) We are launching a weekly podcast about AI in development and beyond. The three of us will lead it: Evgeny Sergeyev, Alexei Litvinov and I, Alexander Polomodov. We have different experiences and different perspectives, but the common interest is to understand how AI is actually changing engineering, products and management. Not at the level of model ratings and magic prompts, but where real codebases, limitations, responsibility and team implementation begin.
- · #Legend
Jeff Dean is leaving Google. 27 years (Category Legend)
Jeff Dean. leaves Google after 27 years (Rubric Legend ) Just a couple of days ago, I was told Jeff gave a lecture to startups from Y Combinator. Towards the end, there was this phrase. Finally, Dean expects that in 2027 Automation of ML itself will increase significantly in the year: the system will lay out the task, run many experiments, evaluate the results and collect an improved version. AlphaEvolve already shows the shape of this cycle. And today, this news comes out, where Jeff and his friends start the "Discovery Loop!" lab in the format.
- · #BookCube
Come, Lesha Litvinov and I started live broadcast about AI in development
Come, Lesha Litvinov and I have started. live AI in development
- · #AI
Materials live with Artem Bondar about how to implement AI in operational work (AI column)
Materials live with Artem Bondar about how to implement AI in operational work (Rubric AI ) Ready materials from the live podcast Code of Leadership with Artyom (@artemonml):
- · #AI4SDLC
Harry Tan on AI-native company: memory is more important than model (Category AI4SDLC)
Harry Tan on AI-native company: memory is more important than model (Rubric AI4SDLC ) Reviewed the report by Harry Tan, President and CEO of Y Combinator,Every company should have a BrainAI Engineer World's Fair 2026. The main thesis is that it is not a special model that gives the advantage, but the organization of work around agents and the accumulated context. According to Tan's own assessment, his programming performance has grown to 400 and, with strict amendments, in 8–80 once. This is not a benchmark, but a personal assessment.
- · #AI4SDLC
Research Insights Made Simple 27AI development as an evolving stack (Category AI4SDLC)
Research Insights Made Simple 27AI development as an evolving stack (Rubric AI4SDLC ) Why does the same model in two coding agents produce so different results? And what should a company really consider its AI stack: model, bandage, tools, data, or the right of an agent to change internal systems?
- · #AI4SDLC
DX Q2 2026: More Code, Less Confidence (Category #AI4SDLC)
DX data from 500+ engineering organizations reveals three gaps: code grows faster than delivery flow, clarity faster than confidence, and time savings faster than product value.
- · #AI4SDLC
FDE: An engineer who needs his own agent (Category AI4SDLC)
FDE: An engineer who needs his own agent (Rubric AI4SDLC ) I saw it. report Vasuman Moza, founder of Varick Agents, with AI Engineer World's Fair Forward Deployed Engineering is usually discussed as a new model of AI implementation. More interesting is what an engineer does with his hands and what tools he needs. And generally. FDE (forward deployed engineer) In this report, it's an engineer built into the client's team. He turns a blurred business problem into a working system: he interviews process owners and restores the real scheme of work.
- · #BookCube
Through 5 Code of Leadership starts live with Artem Bondar about GenAI outside of development:)
Through 5 minutes off live Code of Leadership with Artem Bondar about GenAI out of development:) Come and ask questions, Artem and I will gladly answer them.
- · #AI
Materials live with Evgeny Sergeev about how to build working evals (AI column)
Materials live with Evgeny Sergeev about how to build working evals (Rubric AI ) Ready materials from the live podcast Research Insights Made Simple with Evgeny Sergeev:
- · #Books
Agentic Design Patterns: From Prompt Chains to Agent Systems (Category #Books)
A review of Antonio Gulli's 21 patterns as five levels of complexity, from a controlled model call to an observable, bounded system of agents.
- · #AI
Alexander Wang: If intelligence ceases to be a deficit (AI column)
Alexander Wang: If intelligence ceases to be a deficit (Rubric AI ) Watch Harry Tan Talk to Alexander Wang at Startup School 2026 "Alexandr Wang: “This is a Once-in-a-Civilization Opportunity“, published 29 July 2026 years. Wang founded Scale AI and now holds the position of Chief AI Officer at Meta, which is banned in Russia. But more interesting than posts, his main forecast is that intelligence and the ability to act will become abundant, and the lack of vision, ambition and the ability to assemble the system around agents will remain.
- · #AI4SDLC
ASUS ExpertCenter Pro ET900N G3: local models for $100 thousand (Category AI4SDLC)
ASUS ExpertCenter Pro ET900N G3: local models for $100 thousand (Rubric AI4SDLC ) Looked at the published 30 July 2026 The Alex Ziskind Test of the YearThis was a data center a year ago… Now it's on my desk? It seems that their local models have never been so close. The little thing left: find $100 Thousands on your own desktop supercomputer ASUS ExpertCenter Pro ET900N G3 . And that's hardly an exaggeration. One of the American sellers indicated a price for the system $99,999.99.
- · #SRE
Materials on live TV with Anatoly Krasnovskiy about modeling reliability according to the dependency graph (Category SRE)
Materials on live TV with Anatoly Krasnovskiy about modeling reliability according to the dependency graph (Rubric SRE ) Ready materials from the live podcast Research Insights Made Simple with Anatoly Krasnovskiy (@mb3rlab):
- · #AI4SDLC
IT Picnic 8 August and my post-AI development report and our research (Category AI4SDLC)
IT Picnic 8 August and my post-AI development report and our research (Rubric AI4SDLC ) 8 August will be held in Moscow IT Picnic - summer festival, which T-Bank makes for IT specialists and their loved ones: lecture halls, interactive areas and music on the grass at the Kolomenskoye Museum-Reserve. I’m giving a talk on “Development after AI”: I’ll tell you how AI is now in development – in the industry in general and in T-Bank in particular – and present our AI4SDLC Research study. 2026.
- · #AI
Jeff Dean: The 1% Rule for AI Products (Category AI)
Jeff Dean discusses where small AI teams can compete, context engineering, reliable agent workflows and specialized inference hardware in his YC Startup School 2026 conversation.
- · #DevEx
Code of Leadership S2E8: A Digital Team Lead—Can Developer Effectiveness Be Measured from Code? (Category DevEx)
An August 6 livestream with Ivan Gel about UpCore, its accuracy and effectiveness claims, and the limits of measuring engineering work from code.
- · #AI4SDLC
Sonar and the Verifiers’ Moment (Category AI4SDLC)
Tariq Shaukat on agent reliability, Sonar’s Guide–Generate–Verify–Solve cycle, and why static analysis and clean code matter as agents produce more software.
- · #Architecture
Materials live with Sergey Baranov about AI in architecture (Category Architecture)
Materials live with Sergey Baranov about AI in architecture (Rubric Architecture ) Ready materials from the live podcast Research Insights Made Simple with Sergey Baranova n (@blog\ sb)Scrumtrek Partner and ArchDays Conference Organizer (@blog\ sb):
- · #BookCube
Come to the podcast, we just started it.
Come on. podcastWe're just getting started.
- · #AI
How AI-native development is changing the role of products
How AI-native development is changing the role of products I got myself an interesting thought. The product manager’s routine ends faster than many people notice. In AI-native development, a product ceases to be an interface between “business” and engineers. He needs to go into the data himself, assemble a prototype with the agent, formulate evals and acceptance criteria, understand the cost and risks.
- · #BookCube
At the suggestion of the channel Roots wrote a post about how AI is changing the profession of the product
At the suggestion of the channel Roots I wrote a post about how AI is changing the product profession. Thank you guys for offering to think about it.
- · #AI
Frank Coyle on Ontology: Logic Outside the Probabilistic Agent (AI column)
Frank Coyle on Ontology: Logic Outside the Probabilistic Agent (Rubric AI ) Watch the twenty-minute report by Frank Coyle.Why Agentic Systems Need OntologiesAI Engineer World's Fair 2026. I’ve had a complicated relationship with the word “ontology” for a long time: in conversations about architecture, it was almost a stop word. Too often, this terminology has been used by people who are very far from practice: instead of a working contract, it has been an attempt to classify the whole world first. That seems to be changing now.
- · #Books
AMA Session 2 AI-assisted Engineering with Alexey Litvinov n
AMA Session 2 AI-assisted Engineering with Alexey Litvinov n Wednesday, 17:00 in Moscow together with Alexey Litvinov live We will continue to talk about AI-Assisted Engineering and how to build an AI-Native organization. Za. first series of AMA sessions We didn't manage all the issues, so we will continue to discuss this with Lyosha, who has his own tg channel - @tip\ podcast, subscribe to him.
- · #BookCube
Live's on, come on.
Live's on, come on.
- · #Consulting
Materials live with Alexander Vorontsov about the future of consulting (Category Consulting)
Materials live with Alexander Vorontsov about the future of consulting (Rubric Consulting ) Ready materials from the live podcast Code of Leadership with Alexander Vorontsov, partner of the companyrevelio" (@revelio\ tech.):
- · #SRE
Research Insights Made Simple 25: We model reliability by dependency graph (Category SRE)
Research Insights Made Simple 25: We model reliability by dependency graph (Rubric SRE ) Is it possible to know where a distributed system will break before the expensive crash experiment? In this release tonight 17:00 Together with Anatoly Krasnovskii we will analyze his work.Model Discovery and Graph Simulation: A Lightweight Gateway to Chaos Engineering", marked Distinguished Paper Award on ICSE-NIER 2026 (By the way, I handler before).
- · #Architecture
Why the AI-copilot architect still failed (Category Architecture)
Why the AI-copilot architect still failed (Rubric Architecture ) Starting. live About AI in architecture with Sergey Baranov 5 minutes. Sergey is a practicing architect, Scrumtrek partner and founder of the ArchDays conference, where I have been on the program committee from the first conference to the present. In general, we have known Sergey for a long time and our discussions are usually entertaining:) Come and ask questions during the live broadcast - we will gladly answer them.
- · #Architecture
R2ABench: From Requirements to Architecture—or Why PlantUML Skills Do Not Make an LLM an Architect (Category #Architecture)
R2ABench exposes the gap between identifying architectural components and recovering their relationships, rationale, and traceability to requirements.
- · #Books
Summer Sale at Peter Publishing House ( Books)
Summer Sale at Peter Publishing House ( Books) Traditionally, I'm talking about sales at this publishing house. This time you can order not only on the website of the publisher, but also 🛍 WB and 🛍 ozone. It turns out that these two sites were added in honor of 35- anniversaries of publishing.
- · #Agents
Research Insights Made Simple 26How to Build Working Evals for AI Agents (Agents column)
Research Insights Made Simple 26How to Build Working Evals for AI Agents (Rubric Agents ) How do you know if an AI agent is ready for production? A beautiful response and even a high "pass rate" show only the final of a single launch. The agent could peek at the solution, choose a dangerous path, violate rights or crumble when rerun.
- · #AI4SDLC
State of AI4SDLC on HighLoad++: where development bottlenecks are moving (Category AI4SDLC)
State of AI4SDLC on HighLoad++: where development bottlenecks are moving (Rubric AI4SDLC ) Appeared. record My performance at Saint HighLoad + + 2026, slides and abstract. The report was about what happens when the local acceleration of coding gets into the engineering system of a large company. 10 000+ engineers. The answer is simple: AI speeds up code writing before the organization has time to rebuild the entire supply stream. Therefore, the bottleneck does not disappear, but moves to task setting, review, testing, integration and release.
- · #AI4SDLC
Materials on AMA session with Alexei Litvinov about AI-Assisted Engineering (Category AI4SDLC)
Materials on AMA session with Alexei Litvinov about AI-Assisted Engineering (Rubric AI4SDLC ) Ready materials from the live podcast Code of Leadership with Alexei Litvinov:
- · #Consulting
Code of Leadership S2E7: Consulting in the Age of AI: What is left without beautiful presentations? (Category Consulting)
Code of Leadership S2E7: Consulting in the Age of AI: What is left without beautiful presentations? (Rubric Consulting ) If AI can already gather analytics, make recommendations, and deliver a compelling presentation in a matter of minutes, what will companies pay consultants for? We're starting in. live To discuss this with Alexander Vorontsov, a partner of IT company Revelio Tech: https://t.me/revelio\ tech. Connect!
- · #AI4SDLC
AI Dev Podcast 8Autonomy of the agent begins with limitations. (Category AI4SDLC)
AI Dev Podcast 8Autonomy of the agent begins with limitations. (Rubric AI4SDLC ) Out. The new AI Dev Podcast, in which we, together with Vladimir Yatulchik and Andrei Dmitriev, figured out how to move from vibe coding to managed agent development. TLDR: A standalone agent becomes useful not when it is allowed to write as much code as possible, but when it is embedded in a reproducible engineering process. The more work we delegate to an agent, the clearer the intent, limitations, acceptance criteria, and limits of his rights must be.
- · #Books
If Anyone Builds It, Everyone Dies (If someone creates it, everyone will die.) (Books column)
If Anyone Builds It, Everyone Dies (If someone creates it, everyone will die.) (Rubric Books ) I already am. told I didn't finish reading the book "Agentic Design Patterns" about the creation of AI-agents, as I started "If someone creates it, everyone will die" by Eliezer Yudkovsky and Nate Soares. Now I've finished reading both books, and the contrast is almost comedic: one book explains in detail how to build agents, and the other why, if we build them to superintelligence, we'd better not start at all:)
- · #AI4SDLC
AMA Session on AI-assisted Engineering with Alexey Litvinova n (Category AI4SDLC)
AMA Session on AI-assisted Engineering with Alexey Litvinova n (Rubric AI4SDLC ) Through 5 minute we'll start live Together with Alexei Litvinov, we will talk about AI-Assisted Engineering. We got a run. 15 From how to change an engineer in the age of AI, to how to sell an AI-Native organization to top management. In general, come listen to the answers to the questions and you will have the opportunity to ask clarifying questions along the way.
- · #AI4SDLC
OpenCode Data: What Real-World Usage Reveals About Model Choice (Category #AI4SDLC)
A behavioral snapshot of OpenCode Go: subscription growth, the gap between tokens and users, session economics, cache effects, and an unusual geographic distribution.
- · #AI4SDLC
OpenCode: How Open Boundry Became a Model Market (Category AI4SDLC)
OpenCode: How Open Boundry Became a Model Market (Rubric AI4SDLC ) I saw it. fresh-out YC The Lightcone with Jay V, CEO of OpenCode I'm already in June. handler Talking to Dax Raad about OpenCode: I was more hooked on skepticism about AI hype and the thesis that thinking remains a development bottleneck. The new edition continues the story from a different angle. OpenCode becomes not just another coding agent, but an open layer between developers, models and providers. The scale of Jay describes is almost platform.
- · #AI4SDLC
AI development as a stack: what to rent, adapt and build yourself (Category AI4SDLC)
AI development as a stack: what to rent, adapt and build yourself (Rubric AI4SDLC ) Over the past year, I have been actively engaged in the topic of AI development at the level of a large company. On such a scale, you quickly cease to solve only operational problems: you need to define the target picture, agree on what the engineering system should become, and build a strategy for transition to it. We set goals and strategy as well. But as I moved, I started catching myself with a few less pleasant thoughts.