At Guardian-X, I contribute to the AI engineering behind AZira, a publicly available pest-detection application. I developed and evaluated a FastViT-T12 image-classification pipeline using approximately 18,000 images across 15 pest classes.
As a Machine Learning Intern at FlyRank AI, I built an end-to-end workflow comparing a rule-based ranking approach with a Random Forest model. The Random Forest achieved Precision@50 of 0.74, compared with 0.24 for the baseline.
My projects include CourseFlow, which turns uploaded coursework into evidence-backed requirements and a scheduled plan, and a from-scratch neural network classifier built with NumPy. I’m studying Computer Science at French-Azerbaijani University (UFAZ) and have also built a computer-vision system for recognizing handwritten Azerbaijani characters.
