Massinissa Merouani

Massinissa Merouani

Ph.D. Candidate in Computer Science @New York University
NYUAD Global Ph.D. Fellow

On the Job Market (Available December 2026)

I am a Ph.D. candidate at the NYU Tandon School of Engineering advised by Professor Riyadh Baghdadi. My research interests lie at the intersection of Machine Learning, Computer Systems, and High-Performance Computing.

My research leverages Deep Learning and Large Language Models (LLMs) to solve compiler optimization problems. Specifically, I design systems that explore code transformation spaces to generate highly efficient code for modern hardware. Previously, I was a Research Engineer at NYU Abu Dhabi and a Research Intern at MIT CSAIL (COMMIT group).

Status: I am actively seeking full-time opportunities starting December 2026. I am open to roles in both industry and academia where I can apply my expertise in ML and Systems.

Selected Publications

Nov. 2025 Agentic Auto-Scheduling: An Experimental Study of LLM-Guided Loop Optimization
M. Merouani, I. Kara Bernou, R. Baghdadi
34th International Conference on Parallel Architectures and Compilation Techniques (PACT 2025)
[PDF] [IEEE] [Code]
Nov. 2025 LOOPer: A Learned Automatic Code Optimizer For Polyhedral Compilers
M. Merouani, A. Boudaoud, N. Aouadj, et al.
34th International Conference on Parallel Architectures and Compilation Techniques (PACT 2025)
[PDF] [IEEE] [Code] [Cost Model]
Feb. 2023 A Deep Learning Model for Loop Interchange
L. Mezdour, K. Kadem, M. Merouani, et al.
32nd International Conference on Compiler Construction (CC 2023)
June 2022 A Deep Learning Guided Exploration of Affine Unimodular Loop Transformations
M. Merouani, K. Boudaoud, N. Aouadj, et al.
12th International Workshop on Polyhedral Compilation Techniques (IMPACT 2022)
Mar. 2021 A Deep Learning Based Cost Model for Automatic Code Optimization Outstanding Paper Award
R. Baghdadi, M. Merouani, MH. Leghettas, et al.
4th Conference on Machine Learning and Systems (MLSys 2021)
Also selected for oral presentation at NeurIPS 2020 ML for Systems workshop
[PDF]
LOOPerSet: A Large-Scale Dataset for Data-Driven Compiler Optimization Under review
M. Merouani, A. Boudaoud, R. Baghdadi
Preprint
[PDF] [arXiv] [Dataset]

Experience

Sept. 2022 - Present

Graduate Research Assistant

New York University Abu Dhabi (NYUAD)

Conducting research on AI-driven methods for automatic program optimization. Developing novel approaches using Large Language Models (LLMs) for compiler code generation and optimization.

Jan. 2021 - Aug. 2022

Research Engineer

New York University Abu Dhabi

Worked on data-driven approaches for modeling computer code performance. Contributed to the development of the Tiramisu compiler infrastructure.

Aug. 2019 - July 2020

Research Intern

MIT CSAIL (COMMIT Group)

Worked within the COMMIT group on automatic code optimization. Built deep learning models to guide the selection of efficient code transformations.

Education

Ph.D. in Computer Science

New York University - Tandon School of Engineering

Advisor: Prof. Riyadh Baghdadi
GPA: 4.0/4.0
Recipient of the NYUAD Global Ph.D. Fellowship
Sept. 2022 - Present
Exp. Graduation: Nov. 2026

Master's in Computer Science

Ecole Nationale Supérieure d'Informatique (ESI), Algiers

Major: Computer Systems
Thesis: A Deep Learning Based Cost Model for Automatic Code Optimization in Tiramisu
Sept. 2015 - Dec. 2020

Technical Skills

Languages: Python C/C++ Java SQL
ML & Data: PyTorch TensorFlow Scikit-learn W&B
HPC & Systems: CUDA OpenMP MPI Docker Slurm