Experience
Booking.com, Amsterdam
Worked on machine translation, semantic search for hotels, and LLM-based chatbots. Mainly in Python and Java.
Yandex, Moscow
Worked on many components of Yandex News, including news clustering, summarization, ranking, and recommendations. Python/MapReduce/SQL for prototyping and analytics, C++ for production-ready solutions.
Moscow Institute of Physics and Technology
Taught Algorithms to 1st and 2nd-year undergraduates.
ABBYY, Moscow
Worked in the LingvoLive backend team, improving the search of cards with translations of words. Then worked in the linguistics department on tools for machine learning.
Selected Projects
Multi-agent system for writing scientific papers with language models.
⭐ 46
CodeAct-based agentic framework for autonomous task solving.
⭐ 32
MCP server with tools for scientific research.
⭐ 27
Open datasets and language models for the Russian language. Complete training pipeline for instruction-tuned models.
❤️ 500+ · ⬇️ 2M
Benchmark for evaluating role-playing language models with user emulation.
⭐ 112
Models for abstractive and extractive summarization of Russian texts.
⭐ 174
Russian poetry analyzer and generator using neural networks.
⭐ 177
One of the first contextual morphological analyzers for the Russian language.
⭐ 156
Selected Publications
HotelMatch-LLM: Joint Multi-Task Training of Small and Large Language Models for Efficient Multimodal Hotel Retrieval
Arian Askari, Emmanouil Stergiadis,
Ilya Gusev
, Moran Beladev
Speed Without Sacrifice: Fine-Tuning Language Models with Medusa and Knowledge Distillation in Travel Applications
Daniel Zagyva, Emmanouil Stergiadis, Laurens Van Der Maas, Aleksandra Dokic, Eran Fainman,
Ilya Gusev
, Moran Beladev
PingPong: A Benchmark for Role-Playing Language Models with User Emulation and Multi-Model Evaluation
Ilya Gusev
Do not lose the message while paraphrasing: A study on content preserving style transfer
Nikolay Babakov, David Dale,
Ilya Gusev
, Irina Krotova, Alexander Panchenko
HeadlineCause: A dataset of news headlines for detecting causalities
Ilya Gusev
, Alexey Tikhonov
Dataset for Automatic Summarization of Russian News
Ilya Gusev
Improving part-of-speech tagging via multi-task learning and character-level word representations
Daniil Anastasyev,
Ilya Gusev
, Evgenii Indenbom