Artificial Intelligence Engineer

GROUPE CALCUL
• Le groupe Présentation Le Réseau métier Le RT
• Paysage du calcul
• Actions RT
• Actions réseau
• Nos cafés
• Offres d'emploi
• Suivez-nous Offres d'emploi Artificial Intelligence Engineer Proposée par Mathieu Lobet, le 29 septembre 2026 CDD Maison de la Simulation, CEA Saclay, Gif-sur-Yvette, France CEA 2 ans Email Web Suivez la liste Artificial Intelligence Engineer
• About the position
•

Location:
Maison de la Simulation, CEA Saclay, Gif-sur-Yvette, France
• Contract: Two-year fixed-term contract
• Contact: mathieu.lobet@cea.fr
• Position overview The French Alternative Energies and Atomic Energy Commission (CEA) is seeking a motivated and skilled engineer to work on the European AI Factory project. This position offers the opportunity to contribute to cutting-edge research at the intersection of High-Performance Computing (HPC) and Artificial Intelligence (AI). AI Factory France (aifactory-france.eu) is a large-scale national and European platform supported by a broad coalition of leading partners (GENCI, Inria, CNRS, CEA, France Universités, CINES, Amiad, Mission French Tech, Station F and HubFranceIA). The platform brings together resources from the French ecosystem, drawing on:
• The supercomputing infrastructures operated by GENCI (Jean Zay at IDRIS, Adastra at CINES and Joliot-Curie at TGCC).
• The future Alice Recoque supercomputer, France's exascale system.
• Main responsibilities
• User support and training
• Provide technical expertise and support to users deploying large-scale AI workflows on HPC systems.
• Develop training materials and lead workshops to bridge the knowledge gap in the AI community around the use of HPC.
• Development and optimization
• Contribute to adapting and fine-tuning AI foundation models to meet the needs of European research and industry.
• Develop and run benchmarking methodologies to evaluate AI model performance across different HPC architectures.
• Analyze and document performance metrics to identify bottlenecks and recommend optimizations for scalability and efficiency.
• Collaboration and outreach
• Work closely with international partners to align research activities with project objectives.
• Contribute to writing best-practice guides, technical documentation and scientific publications.
• Represent the project at conferences, meetings and community events.
• Required qualifications Education
• PhD or Master's degree in Computer Science, Artificial Intelligence, Computational Science or a related field, with expertise in HPC or AI. Technical skills
• Excellent Python programming skills.
• Proficiency with AI/ML frameworks (TensorFlow, PyTorch) and their optimization for HPC platforms.
• Knowledge of HPC systems and architectures, especially GPU-based.
• Experience with parallel computing. Other Skills
• Experience with data preparation, large-scale model fine-tuning and deployment on HPC systems.
• Excellent problem-solving, teamwork and communication skills.
• Why join us?
• Contribute to an innovative European project focused on the next generation of scalable AI.
• Work in a multidisciplinary environment with access to state-of-the-art computing resources.
• Gain exposure to strategic international collaborations.
• Salary and benefits Benefits Include CEA offers compensation commensurate with your qualifications and experience.
• 🌍 International collaboration: Work with European, American and Japanese laboratories, with opportunities to travel for conferences and workshops.
• 🏠 Flexibility: Up to 3 days of remote work per week.
• 🚆 Transport: Reimbursement of up to 75% of public transport passes and access to CEA’s dedicated shuttle bus network
• 🏥 Wellbeing: Attractive complementary health insurance and employee savings plans.
• 📅 Leave: 5 weeks of paid leave and 4 weeks of RTT (reduced working time) per year.
•

How to apply
Interested candidates are invited to send their CV and cover letter to: 📧 mathieu.lobet@cea.fr
• Suivez-nous
• Contactez-nous
• Mentions légales Développé avec Pelican / Creative commons Graphisme par pierredigonnet.com