// Profile
A graduate researcher with a deep passion for generative AI, mathematical modeling, and building AI systems that matter.
I'm Zannatul Naim, a Computer Science and Engineering graduate from the University of Rajshahi, Bangladesh (CGPA: 3.68/4.00, Jan 2020 – Dec 2025). My academic journey has been defined by an ever-deepening fascination with the mathematics and engineering behind intelligent systems.
My research centers on generative modeling — specifically how we can use GANs and other generative methods to solve the chronic data scarcity problem in Brain-Computer Interface (BCI) research. During my undergraduate thesis, I implemented a conditional WGAN-GP to synthesize realistic EEG time-series data for SSVEP-based BCI, significantly improving downstream classification performance.
I've published twice at the Undergraduate Conference on Intelligent Computing and Systems (UCICS), including a paper achieving 95.35% cross-subject accuracy on EEG event recognition — a result that required rethinking architectural choices from the ground up.
Beyond research, I'm an avid competitive programmer with 600+ problems solved across Codeforces, UVA, LightOJ, CodeChef, and CSES. This practice sharpens the algorithmic instincts that make me a better researcher and engineer.
I'm currently seeking an MSc position in Generative AI — particularly at the University of Manitoba's CORE AI Lab — where I can contribute to foundational research on large generative models and representation learning at scale.