HSE Develops Its Own MLOps Platform
HSE researchers have developed an MLOps platform called SmartMLOps. It has been created for artificial intelligence researchers who wish to transform their invention into a fully-fledged service. In the future, the platform may host AI assistants to simplify educational processes, provide medical support, offer consultations, and solve a wide range of other tasks. Creators of AI technologies will be able to obtain a ready-to-use service within just a few hours. Utilising HSE’s supercomputer, the service can be launched in just a few clicks.
Many researchers working in AI development lack sufficient experience in building web services, managing distributed computing resources, system administration, information security, and deployment process automation. SmartMLOps was created to help eliminate the need to learn additional technologies and to address issues of repeatability and reproducibility in machine learning experiments.
The project partners include the HSE AI Research Centre, the School of Software Engineering at the HSE Faculty of Computer Science, and the HSE Tikhonov Moscow Institute of Electronics and Mathematics (MIEM). Experienced and highly qualified HSE specialists are involved in the development and deployment of the MLOps platform.
According to the project lead, Hadi Saleh, the platform is already undergoing pilot testing. Its primary goal is to reduce the time and resources required to deliver product-level services. SmartMLOps ensures the necessary level of reliability, security, and transparency in the use of AI technologies. It allows developers to save time on non-core activities such as system administration, DevOps, and CI/CD. The platform provides a unified graphical interface through which users can create a repository with an AI module template, access models, a monitoring system, orchestrators, and the supercomputer, as well as track their services’ metrics and promptly address issues.
To create a multifunctional, user-friendly, and modern service, the platform’s developers tackled the complex challenge of integrating HSE’s cHARISMa supercomputer cluster into cloud computing. The platform is based on a microservice architecture, with the main component being a framework developed by the HSE AI Research Centre.
Hadi Saleh
‘The platform's functionality includes the creation, storage, versioning, training and fine-tuning of AI models; the deployment of ready-made models as product services with REST APIs; the construction of training pipelines; the monitoring of services and the overall state of computing resources; the ability to use data from HSE’s corporate information systems; and access to cloud storage and supercomputer resources. The framework itself and the first AI modules have already been deployed, and analytical dashboards for monitoring the platform’s status have been implemented. A special autoscaling technology was developed and introduced to save resources,’ said Hadi Saleh.
Unlike commercial solutions such as Amazon SageMaker or Google AI, SmartMLOps is primarily aimed at researchers and students at HSE. To gain access, they simply need to submit an application and await approval. Until the end of 2025, anyone interested may take part in the platform's pilot use.
A special ETL process was developed to securely extract and anonymise information from HSE’s corporate systems. This helps in analysing student behaviour and building predictive models, for example, to forecast career prospects.
Elena Kozhina
‘We expect that the implementation of the SmartMLOps system will mark a significant milestone in accelerating the development of AI-based services at our university,’ noted Elena Kozhina, Deputy Vice Rector and project curator.
At present, eight AI models have been deployed on the platform, and integration has been carried out with three of HSE’s corporate information systems. Two subsystems of the MLOps platform, developed with the involvement of specialists from the Faculty of Computer Science and MIEM, have been registered with the Russian Federal Service for Intellectual Property (Rospatent).
Sergey Lebedev
‘The development of such an ambitious and promising infrastructure project posed a considerable challenge and a timely necessity for the university. SmartMLOps enables the simultaneous solution of research, educational, and administrative tasks of various scales and levels. We managed to form a unique team of professionals and involve students and research assistants in the process,’ said Sergey Lebedev, Head of the School of Software Engineering at the HSE Faculty of Computer Science.
Anton Sergeev
Anton Sergeev, Director of the Centre for Software Development and Digital Services at HSE MIEM, emphasised that systematic work with AI models is impossible without MLOps platforms, which support the full lifecycle of models—from data import and cleaning to controlling the outcomes. ‘Such high-tech projects set a benchmark for AI leadership not only within our university, but across the country as a whole. They represent HSE’s scientific excellence in the form of a system that will assist ML specialists in applying AI models to a wide range of real-world problems,’ he said.
Pavel Kostenetskiy
Pavel Kostenetskiy, Head of the Supercomputer Modelling Unit, gave the service high praise. ‘The MLOps platform will make it easier for HSE researchers working in AI to access the ‘cHARISMa’ supercomputing cluster and will lower the entry threshold into high-performance computing. In a sense, HSE’s supercomputer will appear like a cloud within this MLOps platform, which many will find convenient,’ concluded Pavel Kostenetskiy.
See also:
HSE Scientists Develop Method to Compress Large Language Models Without Losing Quality
Researchers from the AI and Digital Science Institute at the HSE Faculty of Computer Science have developed a new compression method for large language models such as GPT and LLaMA that reduces their size by 25–36% without additional training or significant loss of accuracy. This is the first approach to use mathematical transformations—specifically, rotations of model weights—to make models more amenable to compression with structured matrices. The study results have been published in ACL Findings 2025. The code is available on GitHub.
Machine Learning Models Can Help Reduce Volatility and Boost Stock Market Returns
The use of machine learning models makes it possible to achieve greater accuracy in predicting risks in the Russian stock market compared to classical econometric approaches. The predictive power of these models increases by 23%, while the average investor’s return can reach up to 13% per annum. These conclusions were drawn by Nikita Lysenok from the Department of Financial Market Infrastructure at the HSE Faculty of Economic Sciences. The paper has been published in Fundamental and Applied Mathematics.
HSE Study Reveals Imbalance in the Generative AI Market
Researchers at HSE University analysed how effectively the global generative artificial intelligence market converts investment into real revenue, concluding that AI is currently developing faster than it is paying off. The results have been published in the journal Foresight and STI Governance.
HSE Scientists Train Neural Network to 'Hear' Faults in Electric Motors
Researchers at the AI and Digital Science Institute of the HSE Faculty of Computer Science have developed a new method—the Signature-Guided Data Augmentation (SGDA) framework—that achieves 99% accuracy in motor fault detection and 86% accuracy in fault classification. The application of this approach can reduce industrial equipment repair costs, minimise downtime, and improve production safety. The study results have been published in Engineering Applications of Artificial Intelligence.
HSE Graduate’s AI Project Wins at TECH & AI Awards
Daria Davydova, graduate of the HSE Graduate School of Business and Head of the AI Implementation Unit at the Artificial Intelligence Department of Alfa-Bank, received a prize at the TECH & AI Awards. She was awarded for the best AI solution for optimising business processes. The winners were determined as part of the VII Russian Summit and Awards on Digital Transformation (CDO/CDTO Summit & Awards).
New Neural Network for Science and Innovation Being Developed at HSE University
HSE researchers are training large language models (LLMs) to understand Russian-language scientific terminology while improving their energy efficiency. The adapted model runs 2.7 times faster and requires 73% less memory than the original open model, allowing it to operate on more affordable hardware. The programme has passed state registration.
HSE FCS Researchers Showcase AI and Bioinformatics Breakthroughs at ICLR 2026
Researchers from the AI and Digital Science Institute at the HSE Faculty of Computer Science, along with students from the AI360: Artificial Intelligence Engineering track of the Applied Mathematics and Information Science bachelor’s programme, took part in ICLR, one of the world’s most prestigious international conferences on machine learning and representation learning. This year’s event was held in Rio de Janeiro, Brazil.
The Future of Cardiogenetics Lies in Artificial Intelligence
Researchers from the AI and Digital Science Institute at the HSE Faculty of Computer Science have developed a program capable of analysing regions of the human genome that were previously inaccessible for accurate interpretation in genetic testing. The program adapts large generative AI (GenAI) models for cardiogenetics to predict how specific mutations affect the function of individual genes.
HSE and Yandex Propose Method to Speed Up Neural Networks for Image Generation
A team of scientists at HSE FCS and Yandex Research has proposed a method that reduces computational costs and accelerates text-to-image generation in diffusion models without compromising quality. These models currently set the standard for text-to-image generation, but their use is limited by high computational loads, the company said in a statement.
A Trap for the Advanced Student: How to Break the Habit of Blindly Trusting Neural Networks
Andrei Ternikov, Associate Professor at the St Petersburg School of Economics and Management at HSE University–St Petersburg, has developed a method for conducting online exams that significantly limits students’ ability to use ChatGPT and other AI models to obtain correct answers. Andrei Ternikov spoke to the HSE News Service about his approach—which won the HSE University Autumn Educational Innovation Competition, received an Alfa Future grant, and was presented at an international conference in Japan.