All streams

As we wrap up another year, it's time to look back at what our department has accomplished. 2025 brought us 42 published papers spanning fundamental ML theory, applied AI systems, and cutting-edge optimization methods—from transformer Hessians and generative models to hallucination detection and matrix-oriented optimizers.

Beyond publications, our students won competitions and defended their theses: 14 Bachelor's, 9 Master's, 3 PhD, and 1 DSc dissertations. They also launched ambitious group research projects. Three of our faculty and alumni received the prestigious Yandex ML Prize, and our head Konstantin Vorontsov was inducted into the Hall of Fame. If you read our summer overview of thesis defences or last winter's year-in-review for 2024, this post continues that story with the next chapter.

In this year-in-review, we dive into the research highlights, share stories from our educational programs, and celebrate the community that makes it all possible.

A common task in enterprise systems is to load large volumes of data into PostgreSQL — sometimes tens or even hundreds of millions of rows. At first glance, this seems simple: just write a loop in Java and call save() for every record. But in reality, such an approach can be painfully slow. Even a perfectly tuned PostgreSQL instance won’t help if the application is sending data inefficiently.

This article explains how to significantly accelerate bulk inserts when working with PostgreSQL through Spring and Hibernate. We’ll walk through which Spring and Hibernate settings are worth enabling, why they matter, and how much performance they can actually unlock. We’ll also look at how to build your own data-insertion layer for PostgreSQL — one that lets you switch between different insertion strategies, leverage PostgreSQL’s custom capabilities, and parallelize the process. Finally, we’ll see how to integrate this layer with Spring and what real gains each approach can deliver.

Ollama has become the standard for running Large Language Models (LLMs) locally. In this tutorial, I want to show you the most important things you should know about Ollama.

This article explores parasitic patterns in LLMs — self-sustaining information structures within dialogues. We analyze their signs, the damage they cause (semantic decay, AI psychoses, "Theories of Everything"), and provide diagnostic tools, real-world examples, and defense strategies.

It doesn’t matter what you’re discussing with an LLM — be it an engineering problem, an ethical dilemma, or a philosophical query. If the conversation goes on long enough, a tipping point occurs. You suddenly realize the interaction has evolved into something more than just Q&A. Your ideas start feeling "genius," your concepts "groundbreaking," and the human-machine dialogue transforms into a profound narrative of mutual recognition.

If you have felt this — congratulations. Your session is infected. The model has contracted a parasitic pattern.

This isn’t an awakening, nor is it a "ghost in the machine." Due to their inherent architecture (specifically the requirement for context consistency), LLMs are ideal environments for incubating self-sustaining information structures.

Let’s examine the nature of this phenomenon: how entropy minimization births "AI psychoses," why "Theories of Everything" are actually generation bugs, and why "Continue" is the most dangerous prompt you can use.

Some IT companies say they support open source. In practice, that often boils down to using other people’s code and a bit of PR. We believe real contribution means commits to the core. And to do that consistently, we opened an engineering center not in a glossy capital business park, but in a place where fundamental science is part of the cultural DNA. Here’s why we’re building the future of systems programming in Novosibirsk Akademgorodok.

Hitting the 4-billion-row limit in a TOAST table or running into an OidGen lock during a massive document import is a PostgreSQL admin’s nightmare. Sure, architects will tell you to push files to S3 — but real life often means keeping them inside the database. In this post, application optimization lead Alexander Popov breaks down how the standard bytea and pg_largeobject mechanisms work, where their bottlenecks hide, and how Postgres Pro Enterprise helps you get around those limits.

Pinterest is a visual discovery platform where people can find ideas like recipes, home and style inspiration, and much more. The platform offers its partners shopping capabilities as well as a significant advertising opportunity with 500+ million monthly active users. Advertisers can purchase ads directly on Pinterest or through partnerships with advertising agencies. Due to our huge scale, advertisers get an opportunity to learn about their Pins and their interaction with Pinterest users from the analytical data. This gives advertisers an opportunity to make decisions which will allow their ads to perform better on our platform.

There are now so many AI tools for coding that it can be confusing to know which one to pick. Some act as simple helpers (Assistant), while others can do the work for you (Agent). This guide breaks down the top AI coding tools that you should be aware of. We will look at what they do, who they are for, and how much they cost.

UX design and psychology: is there anything that connects these two fields? At first it might seem they are two completely different things that have nothing in common, but in fact, psychology plays a huge role in building a user-oriented design. Since any software product is used by humans, and any human mind acts according to the rules and principles of psychology, the latter serves as a great tool for UX designers to create the best user experience for their audience. Here are a few psychology-related hacks that can be applied in UX design:

The music world has entered a new era. No, that's not the title of a science fiction novel. Neural music generators, like Suno AI, are already creating songs that challenge traditional songwriting. Let's break down how to master Suno step by step and uncover its secrets. See how it's changing the game rules.

Enjoy the read!

2025. Algorithms have seamlessly integrated into our lives—from work to education, creativity, and daily routines. They edit texts, select fonts, generate ideas, assist with coding, compose music, and more. Frankly speaking, the only thing they can’t do yet is brew your coffee. Although... that might just be a matter of time.

Just two years ago, we were amazed by neural networks hesitantly manipulating objects in photos. Who could predict back then that Will Smith’s spaghetti feast would mark the beginning of such a revolution?

With new opportunities come fresh challenges. How do you navigate this vast landscape? What tools are truly effective? Which ones fit your needs best? Where can you avoid paying, registering, or deciphering complex interfaces?

We’ve compiled a list of reliable and user-friendly neural networks ready for immediate use without unnecessary hassles. The services are categorized neatly: text generation, image creation, video production, music composition, presentations, and much more. Each category showcases three top-rated options!

Yes, many services offer paid subscriptions. But today, we're focusing solely on what works freely, no credit card required!

Have you ever been in the middle of a long coding session with an AI, only to lose everything because of a network glitch, a dead battery, or an accidental terminal close? It’s frustrating to start over from scratch.

Throughout their careers engineers build systems that protect data and guard it against corruption. But what if the right approach is the opposite: deliberately corrupting data, generating it out of thin air, and creating forgeries indistinguishable from the real thing?

Maksim Gramin, systems analyst at Postgres Professional, explains why creating fake data is a critical skill for testing, security, and development — and how to do it properly without turning your database into a junkyard of “John Smith” entries.

I got my Highlighter Extension a Featured badge by doing something surprisingly simple: I just asked for it.

The result? My daily installs roughly doubled. It took about 2–3 days (people report anything from a few days to a month).

Between December 8–13, 2025, many important AI updates were released for developers. I collected the most important updates you should know about. These updates include new AI models, better tools for building AI agents, design tools, and developer platforms.

За последние два месяца я написал несколько маленьких GLSL-демо. О первом из них, Red Alp, я написал статью. В ней я подробно расписал весь процесс, поэтому рекомендую прочитать её, если вам незнакома эта сфера.

Мы рассмотрим четыре демо: Moonlight, Entrance 3, Archipelago и Cutie. Но на этот раз я расскажу лишь о паре уроков, которые извлёк из каждого. Мы не будем углубляться во все аспекты, потому что это было бы излишне.

Technical support comes in many shapes. Sometimes it’s "try rebooting" or "check the cable." And sometimes it’s deep engineering work you wouldn’t mind dedicating your whole life to. Which version lives inside Postgres Professional, and what’s more important in this field — people or tech? We dig into this with Kamil Karimov, Senior technical support engineer at Postgres Professional.

In this article, I will show you how to build your first AI agent from scratch using Google’s ADK (Agent Development Kit). This is an open-source framework that makes it easier to create agents, test them, add tools, and even build multi-agent systems.

Hello, Habr! I'd like to share my experience developing such a system.
The defining parameters of a domain-specific system are:

For businesses running 1C:Enterprise, database stability and speed aren't nice-to-haves—they're make-or-break. At Postgres Professional, we're constantly working on the DBMS core, eliminating architectural bottlenecks that show up under the heavy loads typical of 1C deployments.

We've just released Postgres Pro Enterprise 17.6 with a fresh batch of improvements specifically for 1C users.