About Me

Hi, I’m Yurii - an engineering fellow, solution architect, computer scientist, and AI researcher. I specialize in designing, implementing, and testing advanced neural network architectures for real-world applications including damage assessment, healthcare, and navigation.

I mentor research teams, lead R&D initiatives, and consult organizations on AI adoption. I also lecture on Machine Learning and Formal Verification at Taras Shevchenko National University of Kyiv.

Publications

• Research on the state of areas in Ukraine affected by military actions based on remote sensing data and deep learning architectures
• Multiscale Scene Localization Based on Composite Network for Remote Sensing Imageries: A Case Study on Critical Infrastructure, IEEE DESSERT 2024
• The thin bridge between formal methods and medical Explainable AI
• Smarter Regression Testing with LLMs and RAG (No “Agent Army” Required)

Projects

Non-profit projects

Adaptive OSINT Data-Mining Platform NDA

Architected an end-to-end system that ingests multilingual social-media, satellite telemetry, leaks, and other data sources. Leveraged graph mining, cross-lingual embeddings, and transformer-based event extraction to surface actionable threat clusters in near-real time, reducing analyst triage workload by 65 %.

GPS-Spoof-Resilient UAV Navigation NATO SPS Grant

Designed deep-learning models and real-time 3-D reconstruction for a Black-Sea drone fleet, boosting navigation robustness in GPS-denied environments. Delivered sub-10 cm mesh accuracy and filed three internal patents on memory-efficient vision algorithms.

Remote Damage-Assessment System University R&D

End-to-end satellite-imagery pipeline for rapid post-disaster analysis; diffusion-generated synthetic data enlarged training sets and raised detection recall by 12 pp. Adopted by civil-protection agencies for situational awareness.

Conferences