Anar Sabuxi Oğlu Rzayev

pp.png

Hey 👋 I’m Anar Rzayev, a researcher working at the intersection of geometric deep learning and therapeutic discovery. My journey from mathematical olympiads to protein design revealed how proteins, nature’s most complex molecular machines, function through geometric principles that can be understood via mathematics and AI.

At KAIST and EPFL, I develop “context-aware geometric AI” for protein design and analysis. I work with Prof. Ho Min Kim at KAIST, where my research spans from engineering fluorescent proteins to designing antibodies, combining insights from structural biology, medical imaging, and computational geometry.

My research interests include:

  • Geometric Deep Learning: SE(3)-equivariant architectures for molecular design
  • Protein Engineering: Structure-guided design with evolutionary priors
  • Medical AI: Multi-scale molecular-to-clinical translation
  • Diffusion Models: Markov processes for molecular generation

If you’re curious to learn more about my work and experience, you can check out my CV. Feel free to reach out to me via email or connect with me on LinkedIn.

News

Aug 27, 2024 Traveled to Coimbra for Summer School in Computational Biology ✈️
Mar 20, 2024 Rejoining IBS for CRISPR-Cas9 project, exciting developments ahead! 🧬
Sep 20, 2023 Starting my medAI internship at QuantCo 🇩🇪
Mar 01, 2023 Joined Laboratory of Protein Design & Immunoengineering 🇨🇭
Feb 21, 2023 Started my exchange semester at EPFL! 🎉

Selected Publications

  1. structure.jpg
    Structure-based representation for protein functionality prediction using machine learning
    Minji Lee, Anar Rzayev, Hyunkyu Jung, and 3 more authors
    한국정보과학회 학술발표논문집 (Korean Information Science Society Conference Proceedings), Later version presented at ML in Structural Biology Workshop at NeurIPS, 2022
  2. blend.jpg
    BLEnD: A Benchmark for LLMs on Everyday Knowledge in Diverse Cultures and Languages
    Junho Myung, Nayeon Lee, Yi Zhou, and 8 more authors
    arXiv preprint arXiv:2406.09948, Accepted at NeurlPS 2024 Datasets and Benchmarks Track, 2024
  3. worldcuisines.jpg
    WORLDCUISINES: A Massive-Scale Benchmark for Multilingual and Multicultural Visual Question Answering on Global Cuisines
    Genta Indra Winata, Frederikus Hudi, Patrick Amadeus Irawan, and 8 more authors
    arXiv preprint arXiv:2410.12705, 2024