Igor Labutin
AI Researcher, Composer, Producer
Two-time ICPC World Finals contestant and AI safety researcher with leading publications. As a composer, he blends orchestration with rock, pop, and electronics, building a stage debut with an international creative team that fuses science, story, and sound.
Biography
Igor Labutin is a two-time finalist of the ICPC World Programming Championship and an AI Researcher with top scientific publications. His journey began with competitive programming, where he achieved remarkable success including 19th place (out of 58,963 participants) at ICPC 2020 Finals, 62nd place at ICPC 2019 Finals, and 8th place with a silver medal at ICPC 2018-2019 NEERC.
He earned his Bachelor of Science in Computer Science from Saint Petersburg Academic University (2014-2018) and Master of Science in Computer Science from Saint Petersburg Higher School of Economics (2018-2020). His professional experience spans various prestigious institutions, including Yandex, Nvidia, Skolkovo Institute of Science and Technology, and Mayflower where he served as AI RnD Lead.
As a Research Supervisor (Oct 2022 - Sept 2024), Igor supervised research projects for 26 Master and Bachelor students from leading universities including SPb HSE, Moscow HSE, SPb ITMO, SPbSU, Constructor University Bremen, and Neapolis University Paphos. These projects covered diverse AI research areas including Music Generation/Music Information Retrieval, interpretability, and Computer Vision.
As a composer, Igor blends orchestration with rock, pop, and electronics—reshaping genres with precision. A newcomer to the stage world, he's built an international creative team to deliver productions that fuse science, story, and sound. His multidisciplinary approach combines deep technical expertise in AI with creative expression through music.
Research
Current Research Directions
My current research primarily focuses on AI Safety, exploring critical areas that ensure the responsible development and deployment of large language models:
Emotional Triggers for LLMs
Investigating how emotional content and triggers affect language model behavior and responses.
Collaborative Training
Developing methods for training AI systems that better collaborate with humans and other AI agents.
Interpretability
Working on techniques to make AI decision-making processes more transparent and understandable.
Selected Publications
Accelerating regional weather forecasting by super-resolution and data-driven methods
Journal of Inverse and Ill-posed Problems, volume 32, issue 6
2024Enumeration of Labeled and Unlabeled Hamiltonian Cycles in Complete k-Partite Graphs
Journal of Mathematical Sciences, volume 255, issue 1
2021On Some Enumerative Problems in Lambda Calculus
Journal of Mathematical Sciences, volume 247, issue 3
2020MonoForest framework for tree ensemble analysis
Advances in Neural Information Processing Systems (NeurIPS 2019), volume 32
2019Enumeration of 3-regular one-face maps on orientable or non-orientable surface up to all symmetries
arXiv preprint arXiv:1901.06591
2019Productions
Bringing together science, story, and sound
AI: Save our Souls
Edinburgh Fringe 2025
A musical that explores the intersection of artificial intelligence, humanity, and consciousness. This production marks my debut as a producer, uniting an international creative team to deliver a performance that challenges how we understand being human in an age of intelligent machines.
Music
Blending orchestration with rock, pop, and electronics
As a composer, I explore the boundaries between classical orchestration and contemporary genres. My work combines precision with experimentation, creating soundscapes that bridge traditional composition and modern production.
Listen on Streaming Platforms
Contact
Let's connect and collaborate