Publication
Controllable diffusion-based image generation for failure diagnosis of reinforced concrete beam–column joints
Jun 1, 2026 · Journal of Building Engineering
OpenAlex citations: 0Open
DSML Kazakhstan
Personal profile
Yelzhas Omarov
@realyo
Interdisciplinary researcher and developer performing technical coursework and applied research in computer vision and generative models.
Astana city
Featured
A few things that best show this member's work and contribution.
About
Experienced in deep learning model fine-tuning, diffusion-based image generation, and backend system development, with demonstrated research output and teaching experience. Particularly interested in generative modeling and applied machine learning systems.
Career
Experience, skills, projects, and achievements.
Experience: 3 years
AI Researcher and Software Enthusiast
Skills
PythonScikitLearnComputer VisionNLPClassic ML
Experience
Research Assistant
Nazarbayev University School of Engineering and Digital Sciences
Jan 2023 — May 2026- Co-authored a diffusion-based failure diagnosis paper https://www.sciencedirect.com/science/article/abs/pii/S2352710226012878 - Designed a text-to-edge pipeline for beam-column failure visualization tion used in structural diagnostics research - Managed dataset annotation and data augmentation. - Built and evaluated a lung cancer classification pipeline on 1,000 CT images achieved Macro F1=0.9366 and AUC up to 0.9989. - Developed synthetic medical imaging pipelines to improve the model robustness and enables a safer, data-efficient AI deployment
Back End Development Intern
NITEC
Jun 2025 — Sep 2025- Built a secure multi-user Learning Management System (LMS) for government use with JWT/OAuth2 authentication, role-based access control, and granular permissions for students, teachers, and administrators. - Developed a bilingual education platform with Russian/Kazakh localization, collaborative course editing workflows, enrollment management, and rich lecture content delivery with media support. - Engineered a configurable assessment and analytics system featuring automated grading, attempt controls, progress tracking, and performance reporting across courses and students.
Research and publications
A compact preview of this member's public research layer.
Research Assistant
Computer Vision & Generative Models