I explore the nature of optimization — studying algorithms, analyzing data, and trying to solve problems with minimal resources to achieve maximum benefit (in optimal ways)!
"Nature strives for optimization" - physical laws operate on the principle of minimal action, biological organisms are tuned by nature to minimal energy configurations.
Thank you for visiting the webpage, with love @Arman
Research and Development Engineer 🛠️
Supply chain optimization system using LGBM time series forecasting models for predictive analysis of sales and stock levels, enabling proactive warehouse management.
Developed a platform for analyzing and visualizing data on mining incidents.
Developed optimized algorithm combining Dijkstra for minimum pathfinding and time-greedy approach for efficient goods distribution with minimal empty travel time.
Excelled in the "Machine Learning" track of Yandex's prestigious programming and data analysis competition.
Represented Kazakhstan in the international Tech4Good competition, developing innovative solutions for global challenges.
Beyim Insight - AI system for generating detailed SMART reports on student's test/mock exams including feedback, recommendations, visualizations and predictions for various exams (UNT, NUET, IELTS, TOEFL, MESC, SAT).
IEEE AITU 2024: Digital Generation Conference, March 2024
DOI: 10.1109/IEEECONF61558.2024.10585529IEEE AITU 2024: Digital Generation Conference, February 2024
DOI: 10.1109/IEEECONF61558.2024.10585385AITU Digital Generation 2023 Conference, March 2023
DOI: 10.13140/RG.2.2.35223.85927
Neuro-Symbolic ML for Scientific Discovery
OIKAN is a neuro-symbolic machine learning framework inspired by Kolmogorov-Arnold representation theorem. It combines the power of modern neural networks with techniques for extracting clear, interpretable symbolic formulas from data. OIKAN is designed to make machine learning models both accurate and Interpretable.