I am currently working at MSU RCC-LAIR as an NLP Researcher, participating in ruadapt project and other IR research.

• 2021-2025 Bachelor of Computer Science and Mathematics, MSU CMC
• 2025-present time Masters of Computer Science and Mathematics, MSU CMC

• 2024-present time: RusBEIR - comprehensive benchmark designed for zero-shot evaluation of information retrieval (IR) models in the Russian language
• 2024 Coursework: Methods for distilling large language models
• 2024-2025 Diploma: Approaches for LLM distillation and their practical applications
• 2025-present time: WikiFacts-Bench - multilingual Wikipedia-based dataset and framework for RAG benchmark
• 2025-present time: ruadapt - framework for adaptation of Large Language Models to Russian Language

All my datasets/models collected/created for my projects are open-sourced and could be found at HuggingFace

• Building Russian Benchmark for Evaluation of Information Retrieval Models - Dialogue 2025 Conference
• Iterative Layer-wise Distillation for Efficient Compression of Large Language Models - DAMDID/RCDL'2025 (awaiting publication in the conference proceedings)
• Wikipedia-based Datasets in Russian Information Retrieval Benchmark RusBEIR - DAMDID/RCDL'2025 (awaiting publication in the conference proceedings)
• Distillation for Adaptation Language Models to Russian Language (awaiting journal publication)

• HuggingFace: kaengreg
• Telegram: @kaengreg
• Email: kaengreg@ya.ru