Experience

Aug. 2024 - Dec. 2024
GPU/ML Researcher/Developer Intern (part time), Advanced Micro Devices (AMD)
  • Integrated finetuned large language models (Qwen, DeepSeek, Llama) into an agentic AI framework to automate Fortran-to-HIP code translation.
  • Designed a multi agent system using CrewAI, where coordinated agents generated, reviewed, and debugged translated GPU code, achieving higher translation accuracy and performance compared to the zero shot finetuning baseline.
May 2024 - Aug. 2024
Co-Op/Intern: PhD Tech, Advanced Micro Devices (AMD)
  • Finetuned open source LLMs (Qwen, DeepSeek, Llama) to automate Fortran-to-HIP code translation, enabling the generation of a first draft GPU compatible code.
  • Built and curated a dataset, used in finetuning, of input and output pairs that correspond to Fortran code and the equivalent parallel code in HIP.
Jun. 2023 - Present
Graduate Research Assistant, Oden Institute
  • Designed and deployed a high-speed synthetic data generation pipeline for neural operator training, accelerating PDE solutions generation for scientific ML models.
  • Developed a Python library leveraging SymEngine to generate PDE datasets used in training neural operator architectures.

Education

Aug. 2021 - present
University of Texas at Austin (Austin, TX)

PhD Candidate in Mathematics (expected graduation: May 2026)
Research focus: PDEs in operator learning and non-local operators.

Jan. 2020- May 2021
Florida Institute of Technology (Melbourne, FL)

M.S. in Applied Mathematics.
Thesis Tittle: “The Brezis-Nirenberg Problem for the Generalized Kirchhoff Equation”
Advisor: Prof. Kanishka Perera

Aug. 2016 - Dec. 2019
Florida Institute of Technology (Melbourne, FL)

B.S. in Applied Mathematics

Technical Skills

Languages & Libraries
Python (PyTorch, NumPy, JAX, scikit-learn, SymEngine), C/C++
Tools
Git, Linux, Bash

Publications and Preprints

2024
Generating synthetic data for neural operators. Erisa Hasani, Rachel A. Ward. arXiv
2023
On the critical \(p\)-Kirchhoff equation. Erisa Hasani, Kanishka Perera. Topological Methods in Nonlinear Analysis Vol. 61, Issue 1, (Jan 2023) , pgs 383-391. arXiv
2021
On the compactness threshold in the critical Kirchhoff equation. Erisa Hasani, Kanishka Perera. Journal of Discrete and Continuous Dynamical Systems, 2022. arXiv

Summer Schools

Jun. 2023
Princeton Machine Learning Theory Summer School, Princeton University, New Jersey.
Jun. 2022
Summer School in Nonlinear Analysis with a special tribute to Patrizia Pucci, Viterbo University, Italy.

Honors and Scholarships

May 2022
Frank Gerth III Teaching Excellence Award, University of Texas at Austin
Mar. 2017
Faculty Senate Scholarship Award, Florida Institute of Technology
Feb. 2013
Bronze Medal, 30th Greek Mathematical Olympiad (Athens, Greece)