Summary
Khawar Islam is a machine learning and computer vision leader with 11 years of experience, currently serving as CTO for GenAI at Upendi.app and pursuing a PhD in Computer Vision and Machine Learning at the University of Melbourne. He blends deep learning research—spanning vision transformers, self-supervised learning, knowledge distillation and scene graphs—with practical GenAI engineering such as Stable Diffusion refiners, ControlNet, and IP-Adapter deployments. His background includes academic research on abusive content classifiers and large-scale dataset construction (10M+) as well as industrial work on object detection, segmentation, and low-data regimes. Comfortable moving between research and product, he has shipped both foundational models and applied tools for ID/face workflows, OCR, and realistic image synthesis. Notably, he pairs a strong publication and lab pedigree from Sejong University and CVPR Lab with hands-on engineering across PyTorch/TensorFlow and NVIDIA toolchains. He brings a hybrid profile: a researcher who repeatedly turns complex vision concepts into production-ready GenAI features.
10 years of coding experience
5 years of employment as a software developer
Master's degree Computer Engineering, Master's degree Computer Engineering at Sejong University
The University of Melbourne
English, Urdu