Resume

Education

• Ph.D in Artificial Intelligence and Computer Vision, GeorgiaTech-CNRS & CentraleSupelec, 2024 (expected)
• M.S. in Robotics and Machine learning, Paris-Saclay University, 2021
• B.S. Computer Science and Electrical engineering, Paris-Est Marne-La-vallée University, 2019

Work experience

• December 2021 - Present: Research AI at GeorgiaTech-CNRS & CentraleSupelec
  • Consulting in R&D company.
  • Supervision Projects in AI.
  • Teaching ML/DL and Labworks supervision.
• March-September 2021: Data Scientist at Capgemini (Project: Smart Universal Robot (SUR))
  • Using Stacking and Ensemble architecture techniques for detection and estimation of 6D pose of objects (Resnet + EfficientNet V2).
  • Implementation of neural networks for anomalies and defects detection on aircrafts using camera.
  • Train and Inference using DGX server / Deployment On GCP using Docker images.
• Dec 2020 - March 2021: Research Assistant (Project: Anomaly and outlier Detection using Deep Learning)
  • Development of a new neural network architecture.
  • Exploratory data analysis / Preprocessing and features engineering / Classification
  • Train and Inference using Azure Cloud Compute instance.
• Sept 2020 - Dec 2020: Research Assistant (Project: Transfer Learning for EEG signals in BCI)
  • Data cleaning / Features engineering / Preprocessing \hfill
  • Implementation of pipelines and models of Machine Learning (Random Forest, SVM) for BCI.
  • Presentation of published article at IEEE NER 2021 conference

Skills

• General: Problem-solving proficiency, Ability to work independentyly, Advanced synthesis and characterization, Deep-dive data analysis.
• Programming language: Python, C/C++, Bash, SQL, NoSQL, Hadoop, PySpark, RAPIDS, Scala, SAS.
• Topics: Data Science / Engineer, Machine Learning, Deep Learning, Computer Vision, NLP, Cloud Computing.
• Dev tools: Pytorch, Scikit-learn, Tensorflow, Keras, Fastai, Git, Linux, Power BI, Docker, Kubernetes, Cuda, DGX.
• Cloud architecture: Google Cloud Plateform (GCP), Azure Cloud.

Publications

Minimizing subject-dependent calibration for BCI with Riemannian transfer learning

S. Khazem, S. Chevallier, Q. Barthélemy, K. Haroun and C. Noûs, "Minimizing Subject-dependent Calibration for BCI with Riemannian Transfer Learning," 2021 10th International IEEE/EMBS Conference on Neural Engineering (NER), 2021, pp. 523-526, doi: 10.1109/NER49283.2021.9441279

Deep learning for the detection of semantic features in tree X-ray CT scans

KHAZEM, Salim, RICHARD, Antoine, FIX, Jeremy, et al. Deep learning for the detection of semantic features in tree X-ray CT scans. Artificial Intelligence in Agriculture, 2023, vol. 7, p. 13-26, doi: 10.1016/j.aiia.2022.12.001

Improving Knot Prediction in Wood Logs with Longitudinal Feature Propagation

KHAZEM, Salim, FIX, Jeremy, et PRADALIER, Cedric. Improving Knot Prediction in Wood Logs with Longitudinal Feature Propagation. arXiv preprint arXiv:2308.11291, 2023.

The largest EEG-based BCI reproducibility study for open science: the MOABB benchmark

CHEVALLIER, Sylvain, CARRARA, Igor, ARISTIMUNHA, Bruno, et al. The largest EEG-based BCI reproducibility study for open science: the MOABB benchmark. arXiv preprint arXiv:2404.15319, 2024.

PoTATO: A Dataset for Analyzing Polarimetric Traces of Afloat Trash Objects

Luis Felipe Wolf Batista, Salim Khazem, Mehran Adibi, Seth Hutchinson, Cedric Pradalier. PoTATO: A Dataset for Analyzing Polarimetric Traces of Afloat Trash Objects. arXiv preprint arXiv:2308.11291, 2023.

Teaching

Machine learning

Natural Language Processing