cv

IMOLA: Automated Road Network Extraction & Topological Mapping

sponsor: NSF
advisor: Dr. Yao-Yi Chiang
status: ongoing

• Developing deep-learning methods for extracting road networks from maps and building topology reconstruction algorithms

CriticalMAAS: Legend Detection & Map Layout Analysis

sponsor: DARPA CriticalMAAS
advisor: Dr. Yao-Yi Chiang
status: completed

• Developed deep-learning methods for legend detection, achieving 96% symbol-detection F1 and 97.1% symbol-description linking accuracy

Methods for Detecting Critical Events in System Log Data

advisors: Oleg Gorokhov, Mikhail Petrovsky

• Developed NLP-inspired preprocessing and semantic extraction techniques for unstructured system logs
• Built CNN-based anomaly detectors achieving 98% accuracy on processed logs and 82% on raw logs
• Proposed process-mining-based clustering for modeling log-event sequences at scale

Monte Carlo Algorithms for Connect6

advisor: Olga Oparina

• Benchmarked heuristic algorithms for stochastic gameplay modeling and showed Monte Carlo methods outperform classical heuristics

Ph.D.

University of Minnesota Twin Cities

Computer Science

B.Sc.

Lomonosov Moscow State University

Computer Science

Detecting Legend Items on Historical Maps Using GPT-4o with In-Context Learning

GeoSearch '25, ACM SIGSPATIAL Workshops

DIGMAPPER: A Modular System for Automated Geologic Map Digitization

arXiv preprint arXiv:2506.16006

Context-Aware Trajectory Anomaly Detection

1st ACM SIGSPATIAL International Workshop on Geospatial Anomaly Detection (GeoAnomalies'24)

Detection of Anomalies in System Log Data Using Process Analytics Methods

Scientific Conference "Lomonosov Readings - 2023"

Methods for Detecting Critical Events in System Log Data

Scientific Conference "Tikhonov's Readings - 2022"

Evaluation and Comparison of Algorithms for the Connect6 Game with Monte Carlo Method

Journal of the LXX Open international Student's Scientific Fair of Moscow Polytechnic University

Machine Learning Researcher

Rotec Digital Solutions, Moscow, Russia

Developed ML models for anomaly detection in industrial sensor and time-series data for predictive maintenance

• Integrated models into monitoring pipelines using PyTorch, scikit-learn, PostgreSQL, ClickHouse, MLflow, DVC, Airflow, Optuna

Full-stack Java Developer

MOLNET, Moscow, Russia

Implemented backend systems using Java, Spring, JSP, Maven, REST APIs, and PostgreSQL

• Built web interfaces with React, TypeScript, Bootstrap; used Docker, Git, and Linux in daily development

Freelance Writer

JetBrains Academy, remote

Authored educational modules on probability, statistics, data structures, and algorithms for online CS courses

TV NeuroTech hackathon. 1st prize

Case: to find, formalize and explore the statistical validity of the dependencies of indicators of the quality of life of the population (income, demography, credit burden, etc.) on the indicators of state resource provision | Data: official statistics of the Russian Federation for the last 10-15 years | My task was to detect and clean outliers in the data.