Salman Farhat

Avenue Paul Langevin, Villeneuve D'ascq

A motivated researcher with problem-solving and critical thinking skills, as well as meticulous attention to details and methodical nature.
Currently, I'm doing a Ph.D. at Inria laboratory on a subject entitled Safe Dynamic Reconfiguration of cloud applications




The goal of this research project is to design a novel framework for safe dynamic reconfiguration of cloud applications by integrating into the platform software and extending the coordination mechanisms proposed by JavaBIP.

Oct 2020 - Present

Research Intern

LIG - Grenoble Informatics Laboratory

Distributed Learning in the Presence of Byzantine Faults:

  • Study the state-of-the-art in the fields of Distributed Learning Systems and Byzantine-resilient Systems
  • Assess the impact of Byzantine failures on Distributed Learning Systems
  • Propose algorithms and techniques to deal with Byzantine Failures in Distributed Learning Systems
  • Implement and evaluate the proposed algorithms

Feb 2020 - Sep 2020

Research Intern

LIG - Grenoble Informatics Laboratory

Ipanema is a domain-specific language to define schedulers of processes, and the characteristics for process scheduling on a multi-core system. Ipanema is based on Bossa, a language allowing to write safe user-defined scheduling policies on single-core systems.

Jan 2019 - Jul 2019

Full Stack Web Trainee

SE Factory

SE Factory is an intensive full-stack web development Bootcamp that has been running since February 2016 in Beirut, Lebanon

Jul 2018 - Sep 2018


Lebanese University

Software that detects the ID number of the Lebanese University ID card to manage attendance. It was implemented using Python language

Feb 2018 - Jun 2018


University of Lille

Safe dynamic reconfiguration of cloud applications
2020 - Present

Université Grenoble Alpes

Master's degree
Computer science - Parallel and Distributed Systems
2018 - 2020

Lebanese University

Bachelor degree
Computer Science
2015 - 2018


Design a model transformation to transform feature models (static model) into a component-based variability model (runtime model). Information about the design can be found here

Implementation of the model transformation can be found here