Zeyneb Kaya

Zeyneb Kaya is a Co-founder at Topological, a Y Combinator-backed startup in the S25 cohort, and is based in Saratoga, California. She studies computer science, physics, and mathematics at Stanford University and was named a 2024 Regeneron Science Talent Search finalist for her early natural language processing research. Kaya developed MADLIBS, a multilingual data augmentation algorithm designed to improve neural machine translation for low-resource and endangered languages, yielding a 2023 co-authored paper with seven citations. She previously created an artificial intelligence model and mobile application for plant disease identification that has accumulated thousands of downloads. In 2025, she co-authored published research exploring the unified cognitive consciousness theory for language models, securing five citations shortly after its release. Her current focus centers on advancing efficient machine learning algorithms, synthetic data generation, and pretraining methodologies within the artificial intelligence sector.