AI Installation
Université de Genève
GPU pipeline set up for Université de Genève, enabling teams to easily load local AI models for their research and internal applications.
The Challenge
The Université de Genève teams needed to run artificial intelligence models locally for their research work and internal applications, without depending on external cloud services. The absence of dedicated infrastructure and a structured pipeline made loading and using AI models complex and hard to access for the teams, slowing the adoption of these tools in their daily work.
Our Approach
AmenityDev accompanied the Université de Genève in setting up a GPU pipeline designed to simplify the loading and execution of AI models locally. The solution allows teams to easily access the models they need for their research and internal applications, without going through cloud services. The accompaniment included infrastructure configuration, model deployment, and team onboarding.
Results
- GPU pipeline operational, enabling teams to load local AI models independently.
- Model access simplified for research teams and internal applications alike.
- No cloud dependency for model execution, ensuring data sovereignty and confidentiality.
- Teams accompanied and trained to use the pipeline independently.