- 🎓 Current Status: Investigating complex neural phenomena at the University of Texas at El Paso (UTEP).
- 🧪 Primary Research: Deep Learning models.
- 🌱 Recent Breakthrough: Published IMPUGAN at IEEE BigData 2025 convincing GANs to fix broken data.
- 👨💻 Legacy Artifacts: Three years of industrial-scale engineering at WellDev (2022-2025), surviving the transition from Rails monoliths to PyTorch tensors.
- 💬 Ask me about: PyTorch, TensorFlow, and why my Valorant rank is inversely proportional to my research output.
- 📫 How to reach me: zalishmahmud1@gmail.com.
- 🌐 Website: zalish.vercel.app
Research & Deep Learning
Infrastructure & Engineering
Legacy / Industrial Stack
- IMPUGAN (IEEE BigData 2025): Robust data imputation via conditional generative models.
- Explainable Deepfake Detection: Identifying generative lies using CapsuleNets.
- Cyberbullying Detection: NLP analysis of harassment patterns in Bangla.
- Bangla Text Dataset: Exploratory analysis and dataset creation for harassment detection.
"Turning coffee and GPU hours into research drafts since 2025."