All streams

The students of the Intelligent Systems Department successfully defended their bachelor’s and master’s theses. This year, 14 Bachelor’s and 8 Master’s students earned their degrees in Physics, Mathematics, and Computer Sciences. We are proud to say that our Department is unique in publishing the complete set of defense materials during the last ten years. These materials include the text of the dissertation work, the published papers, the code of the computational experiments, and the slides with video of the defense talk.

In this post, we gladly summarize the defended works of our BS and MS students and highlight the results. A recording of their pre-defence presentations can be found here and here in Russian. Most part of the theses has a publicly available English version. 

An Integrated Approach: Philosophical Barriers to Creation and Practical Methods for Detecting Subjectivity in Artificial Systems (from modeling AI consciousness to diagnosing it)

Complex systems often appear chaotic or incomprehensible, yet closer examination reveals that such complexity can frequently be reduced to a simple underlying mechanism. By systematically removing layers of emergent behavior, one can uncover a fundamental rule or equation from which the entire system originates.

For years, we’ve studied Oracle to make PostgreSQL a more migration-friendly option. We introduced tools similar to SQL profile and SQL plan baseline as AQO and sr_plan extensions. In some cases, PostgreSQL even outperforms Oracle, especially in automatic re-optimization.

Migrations from Oracle to PostgreSQL are usually smooth performance-wise, and we’ve even developed session variable extensions to make the transition easier. While many enterprise-only features exist, PostgreSQL often integrates popular solutions directly into the core.

In this tutorial, I’ll explain in simple terms what AI, AI agents, and workflows are, and then I’ll walk you through building your very first AI agent in Python using Google’s Agent Development Kit (ADK). By the end, you’ll understand the differences between these concepts and have a working content-assistant agent you can run from your terminal or a web interface.

Admit it, how many times have you wanted to quickly create an image for a post or presentation, but instead got stuck in an editor or endless searches for a suitable image on Google? Wouldn't it be great if the picture in your head could just appear instantly? Time is money, inspiration is on pause, and that's where AI comes to the rescue. Neural networks can generate anything you want, including the craziest ideas. No need to spend hours searching when, with a few clicks, you can see what was in your thoughts just a second ago.

By the way, notice the cover with the dinosaur? Let's call him Rex. Rex is himself a product of neural network creation. Today he'll be the main star of our experiments. But what will we do? Remember I mentioned crazy ideas? Well, to understand all the possibilities of generation, let's give AI a difficult task. We'll send Rex somewhere in space, for example to the Moon, let him put on a spacesuit and and have him grill some barbecue with Earth in the background. Interested? Then buckle up, we're heading into the world of image generation.

Historically, we’ve ended up using a few rather obscure PHP extensions—written and barely (if at all) maintained by their original authors—that aren’t available in standard Debian package sources.

We stick to the principle of “do it right, and it’ll work right,” which means Slackware-style dropping binaries into the system outside of package managers is frowned upon.

So instead, we’ll be building proper .deb packages for PHP extensions—without breaking compatibility with the existing environment.

A lot of people around me spend time trading on the stock market. Some trade crypto, some trade stocks, others trade currencies. Some call themselves investors, others call themselves traders. I often see random passersby in various cities and countries checking their trading terminals on their phones or laptops. And at night I sometimes write analytical or backtesting software—well, I did up until recently. All these people share a common faith and a set of misconceptions about the market.

We are a brokerage platform operating in a dynamic and complex domain. This specificity comes with a set of challenges. On the one hand, it entails a high variability of scenarios and potentially significant risks associated with errors. On the other hand, it has short development iterations with frequent delivery cycles. 

In this article, we will share how we maintain the quality of our numerous backend services, which provide essential information to our trading terminals.

The goal of the Intelligent Systems Department is to facilitate the road to high-quality professional life. The Ph.D. degree requires three publications in peer-reviewed journals. They are the core of the student’s thesis. This year each of our bachelor students delivered at least one publication. It means they pave the road to their Ph.D. To facilitate this, the Department provides state-of-the-art research topics, scientific advisors with excellence in science, and fine-tuned educational courses. Below, we are proud to recognize our students for their outstanding achievements.

Gen Z – a generation of young people born between the last few years of the 20th century and the first ten years of the 21st, are the primary users of a modern-day Internet. They started using technology since their early childhood, being almost constantly glued to their smartphones and tablets, which led to them being very demanding users that have their own requirements in UX. At the same time, they are known for their sense of humor and straightforwardness – which means they love using products that are both easy to navigate but full of interactive elements. Below we’ve collected a few things you should consider when creating a UX design aimed at a young audience.

Why might a feature not get to production on time? What prevents the team from releasing before the deadline? Can the product owner influence the timeline?

In this article, I will try to answer these questions and suggest solutions for the processes we are implementing and improving within the product team at EXANTE.

SSH (Secure Shell) is the backbone of remote system administration and secure remote access, serving millions of developers and system administrators daily. However, when SSH connections fail, the cryptographic nature of the protocol can make debugging challenging. The complex interplay between authentication mechanisms, encryption algorithms, and network layers often obscures the root cause of connection issues. This complexity is further compounded by the protocol's security-first design, where error messages are intentionally vague to prevent potential attackers from gathering system information. Whether we're dealing with key authentication failures, network connectivity issues, or configuration mismatches, understanding the underlying SSH architecture becomes critical for effective troubleshooting.

In today's interconnected development world, secure authentication is not just a luxury—it's a necessity. Whether you're a seasoned DevOps engineer or a junior developer just starting your journey, understanding SSH key pairs is crucial for your daily workflow. They're the unsung heroes that keep our git pushes secure, our server access protected, and our deployments safe from prying eyes.

But let's be honest: SSH keys canbe confusing. With terms like “public key infrastructure,” “cryptographic algorithms,” and “key fingerprints” floating around, it's easy to feel overwhelmed. This guide aims to demystify SSH key pairs, breaking down complex concepts into digestible pieces that will help you make informed decisions about your security setup.

This article is the first in the series about the upcoming PostgreSQL 18 release. Let us take a look at the features introduced in the July CommitFest.

Planner: Hash Right Semi Join support
Planner: materializing an internal row set for parallel nested loop join
Planner support functions for generate_series
EXPLAIN (analyze): statistics for Parallel Bitmap Heap Scan node workers
Functions min and max for composite types
Parameter names for regexp* functions
Debug mode in pgbench
pg_get_backend_memory_contexts: column path instead of parent, new column type
Function pg_get_acl
pg_upgrade: pg_dump optimization
Predefined role pg_signal_autovacuum_worker

Firebase Performance offers a free, comprehensive solution for tracking app performance. As part of the Firebase suite, it provides seamless integration with other Firebase services like Crashlytics, making it easier to manage all performance and crash data in one place. This not only simplifies access but also streamlines team collaboration without the need to manage multiple platforms.

Recently I have bought starfive visionfive-2 SoC for my own experiments, honestly speaking
I am striving to work with risc-v. After some time I decided to share my experience. Here my bulletpoints:

Small preparation USB-to-Serial connector Write image to microSD/SSD Set boot mode settings

Boot Update bootloader Build kernel Native build Pod build Cross-build on amd64: fast and handy

Chroot to risc-v system from amd64 and install packages

Bonus 1: run qemu with risc-v

Bonus 2: build deb packages for risc-v

Bonus 3: kernel build script

Conclusions

I came across a script on the Internet that allows you to parse product cards from Amazon. And I just needed a solution to a problem like that.

I wracked my brain while looking for a way to parse product cards from Amazon. The problem is that Amazon uses different design options for different outputs, in particular – if you need to parse the cards with the search query "bags" – the cards will be arranged vertically, as I need it, but if you take, for example, "t-shirts" – then the cards will be arranged horizontally, and in such way the script falls into an error, it works out opening the page, but does not want to scroll.

A story about evolving our CI process for a monolithic Python repo with tests, challenges faced, and solutions. Discussing Docker, linters, Allure TestOps, and more.

I’ve been wanting to sort it out about String memory optimization and all these ways to improve the performance and memory usage for Collections and Arrays in C#(as you remember String is an array of chars, loaded a bit differently but anyway) code. I finally managed to find some time to dive deeper into the System.Span.

I have put together this guide to share what I’ve learned. It’s filled with practical tips and examples to help you leverage Spans in your own projects. If you want to optimize your C# code, this guide is a great place to start!