Open to ML & Data opportunities
I build and evaluate practical systems across deep learning, NLP, recommendation systems, and computer vision.
NUMPY / DEEP LEARNING / API
A two-layer neural network implemented from scratch with NumPy: forward propagation, backpropagation, L2 regularization, dropout, and numerical gradient checking.
The model is exposed through tested API endpoints and packaged with Docker.
COMPUTER VISION / BROWSER ML
A browser-based boxing practice application using on-device pose estimation, temporal attempt capture, and rule-based technique feedback for three exercises across two stances.
Completed-score coverage was 55/100 clips; cross was the strongest exercise at 71.8% F1.
NLP / RECOMMENDATION SYSTEM
A hybrid ranking baseline combining semantic similarity from Sentence Transformer embeddings with categorical interest overlap over multilingual student-club data.
Recall@5 and NDCG@5 evaluation remains planned because reviewed relevance labels are not available yet.
I study Computer Science at Nazarbayev University and use personal projects to understand ML beyond a notebook: data preparation, model logic, evaluation, interfaces, testing, and delivery.
B.S. Computer Science · 2024—2028
Neural Networks and Deep Learning ↗
nFactorial SchoolData Science · Silver Certificate ↗
Strengthening model evaluation, deep-learning fundamentals, and reproducible ML delivery.
03 / CONTACT