Nikita Kazeev

Curriculum Vitae

Nikita Kazeev

Download PDFJan 2026

Research scientist at the intersection of AI and physics. Postdoc with Kostya Novoselov at NUS.

12

years in research

9

students mentored

$3.4M

grant co-PI

20+

conference talks

Breakthrough Prize logo

Breakthrough Prize 2025

Skills

Work Experience

National University of Singapore

Postdoc under Kostya Novoselov

2022–present

Leadership

AI4X 2025 conference, ICLR 2025 workshop

  • Speaker search and selection for the 500-person AI4X 2025 conference.
  • Main organizer of the ICLR 2025 workshop on multiscale machine learning.
Transformer for generating symmetric crystals

ICML 2025, code

  • Created a generative model for material design based on symmetry inductive bias.
  • Implemented the model in PyTorch.
  • Achieved the best generation quality and diversity among space-group-conditioned models.
  • Led a team of 6.
ML for predicting properties of defects in 2D crystals

paper 1, paper 2, patent, conference, code

  • Created a sparse representation of crystals with defects, achieving 4x energy prediction error reduction.
  • Developed code for reproducible parallel running of machine learning experiments.
  • Led a team of 3.

CERN [Yandex → HSE University]

Researcher

2014–2022

2025 Breakthrough Prize in Fundamental Physics as a member of LHCb.

Generative models uncertainty estimation

paper, conference, code 1, code 2

  • Developed the first methods for estimating uncertainty of conditional GANs.
  • Led a team of 3 students.
Generative models for fast simulation

paper 1, paper 2, paper 3, code, news 1, news 2, news 3, news 4, news 5

  • Developed a GAN-based machine learning model for high-fidelity simulation of a Cherenkov detector trained on tabular data.
  • Achieved approximately 10^5 speed-up compared to the ab-initio simulator.
Machine learning on noisy data

paper 1, paper 2, code, presentation video

  • Developed a rigorous way to train ML algorithms on data with the label-noise model common in high-energy physics.
Muon identification at the LHCb experiment at CERN

paper, poster, IDAO-2019

  • Solved a classification problem over tabular data with noisy labels under timing constraints.
  • Developed a gradient boosting model that reduced the false positive rate by 30% in the critical low-track-momentum region.
  • Integrated the model into LHCb production, mostly in C++.
  • Packaged the work as a data science competition problem for IDAO-2019.
CatBoost aka fighting biases with dynamic boosting

paper

  • Set up a distributed system for running experiments for the team with Bash and Python.
  • Studied gradient boosting improvements with experiments on toy data.

Education

Higher School of Economics (HSE University)

2016–2020

PhD in Computer Science, supervisor Andrey Ustyuzhanin; Machine Learning for particle identification in the LHCb detector.

Sapienza — Universita di Roma

2016–2020

PhD in Physics (double degree with HSE), supervisor Barbara Sciascia.

Product Management course at Yandex

Spring 2018

Focused coursework in product strategy and execution.

Yandex School of Data Analysis

2013–2015

Master's level CS course covering algorithms, machine learning, deep learning, and distributed systems.

Moscow Institute of Physics and Technology

MS in Physics (2014–2016): Optimisation of data processing of the LHCb experiment.
BS in Physics (2010–2014): Study of the quantum states of the electrons in nonideal plasma and selected molecules using wave packet molecular dynamics with packet splitting.

Mentorship

  • Mentored 9 students (1, 2, 3, 4, 5, 6, 7, 8, 9), 3 interns, and student workshop projects (1, 2).
  • Student council member from 2013 to 2016, leading several dormitory-scale IT projects.
  • Co-PI of a $3.4m AI Singapore grant on multiscale machine learning.

Teaching & Outreach

Selected Publications

Full list on Google Scholar →
  1. Kazeev, Nikita, et al. WyckoffTransformer: Generation of Symmetric Crystals. ICML 2025.
  2. Kazeev, N., Al-Maeeni, A.R., Romanov, I. et al. Sparse representation for machine learning the properties of defects in 2D materials. npj Comput Mater 9, 113 (2023).
  3. Anderlini, L., Chimpoesh, C., Kazeev, N., Shishigina, A., & LHCb collaboration. Generative models uncertainty estimation. Journal of Physics: Conference Series, 2023. I'm the corresponding author.
  4. M. Borisyak and N. Kazeev. Machine Learning on data with sPlot background subtraction. Journal of Instrumentation 14.08 (2019). I'm the corresponding author.
  5. Derkach, D., Kazeev, N., Ratnikov, F., Ustyuzhanin, A., & Volokhova, A. (2019). Cherenkov detectors fast simulation using neural networks. Nuclear Instruments and Methods in Physics Research Section A. I'm the corresponding author.

Service

  • Reviewer for RSC Advances, Machine Learning: Science and Technology, and AI for Accelerated Materials Design workshops at NeurIPS and ICLR (2023–2025).
  • Peer Staff Supporter at NUS, serving as a first-line support for mental wellbeing.

Also

  • Aspiring stand-up comedian; bombed more than 20 times at open mics.
  • International Junior Science Olympiad 2008 in Korea, silver medal.