Our theoretical research is focused on developing a new scientific apparatus for the analysis and design of intelligent systems. Our work involves the formal description of knowledge dynamics—its creation, transmission, and transformation—and the systematic study of "technologization," the process by which cognitive functions are externalized into technology. The goal is to create a rigorous and coherent theory that provides a new language for describing intelligent processes and serves as a blueprint for engineering the next generation of AI.

Formalizing the Dynamics of Knowledge in Intelligent Systems
This project aims to develop a formal ontology of knowledge work, creating a precise language to describe the lifecycle of knowledge from data to application. By defining the operations and transformations involved, we seek to establish a new conceptual framework that enables more robust modeling and design of intelligent systems.

Technologization and the Reproduction: Models and Principles
This research investigates "technologization" as the core process of translating tacit, intuitive knowledge into formal, alienable structures. We analyze this process as both a fundamental bottleneck and a necessary precondition for the creation of intelligent systems. The project's goal is to build formal models of technologization and the reproduction of activity to identify principles that can overcome current barriers and lay the groundwork for designing the next generation of AI.

Our applied research is focused on testing theoretical concepts in practice and developing solutions for current challenges in the AI industry. A key objective of our work is to remove critical barriers to the application and development of intelligent systems, thereby accelerating their adoption and impact. We have identified the following strategic priorities:

Properties and Industrial Application of LLMs. We investigate the fundamental properties and limitations of LLM-based systems to address challenges in their industrial application. The goal is to develop robust architectures and solutions for deploying generative AI in enterprise-level systems.

Data and Knowledge Architectures. We research and develop novel approaches and tools for structuring and managing the complex data and knowledge resources required for advanced generative AI solutions.

Next-Generation Human-Computer Interaction (HCI). We study and design new paradigms for AI-centric user experiences, exploring how human interaction with technology fundamentally changes when mediated by intelligent systems.

Autonomous Agents and Multi-Agent Systems. We research and develop frameworks for the construction of autonomous agents and the systems that govern their interaction, focusing on collaborative problem-solving and emergent intelligence.