AI Solutions Architect: Neural Net Evaluation & LLMs

NVIDIA; Marseille; Sur place; EUR 110 000 - 150 000; Plein temps

Résumé du poste

NVIDIA is seeking an AI-focused Solution Architect to evaluate neural networks and agentic pipelines for enterprise customers across EMEA. You will drive demonstrations and work with customers to tailor NVIDIA technologies to their needs.

As a key technical liaison, you will collaborate with engineering, product and sales teams to promote robust model evaluation methods and integrate LLM/VLM technologies into end-to-end workflows.

Qualifications

• MS/PhD or equivalent in CS, data science, engineering, physics, mathematics
• 5+ years hands-on experience in building, evaluating deep neural networks
• Proven ML/DL/DS experience with emphasis on deep learning
• Experience with latest LLMs or VLMs and retrieval systems
• Understanding of evaluation libraries such as Eval Harness and NVIDIA Evaluation Microservice
• Knowledge of modern evaluation benchmarks and ability to extend them
• Excellent verbal and written English communication

Responsabilités

• Work with key customers to understand their tech and provide best solutions.
• Drive adoption of NVIDIA evaluation tools and establish standard methodologies.
• Develop and demonstrate solutions based on LLM/VLM tech and integrate into agentic pipelines.
• Collaborate with engineering, product and sales to tailor solutions and gather feedback.
• Build industry expertise and integrate NVIDIA tech into enterprise architectures.

Connaissances

Deep Neural Networks

LLMs

VLMs

Retrieval Systems

English communication

Formation

MS/PhD or equivalent experience

Outils

Eval Harness

NVIDIA Evaluation Microservice

Description du poste

NVIDIA is seeking an AI-focused Solution Architect to evaluate neural networks and agentic pipelines for enterprise customers across EMEA. You will drive demonstrations and work with customers to tailor NVIDIA technologies to their needs.

As a key technical liaison, you will collaborate with engineering, product and sales teams to promote robust model evaluation methods and integrate LLM/VLM technologies into end-to-end workflows.