Solutions Architect ยท Paris
AI infrastructure,
made practical.
I design the path from AI use case to production, across Kubernetes, GPU inference, hybrid edge architectures and secure deployment.
Selected work
Case studies
01Putting four AI use cases into productionA team-built EKS platform with GitOps, workload-level cost attribution and an Apache Iceberg data layer. My work covered both the platform and the production onboarding of four use cases.02Keeping NVIDIA vision workloads at the edgeA cloud-designed NVIDIA vision stack reworked to run on Jetson hardware on-premises, so GDPR-sensitive video is processed in the store and only analytics reach AWS.03Building the DevOps foundation for a customer-facing agentDevelopment and production environments built from scratch, with CI/CD, a reusable Terraform deployment and keyless OIDC federation into AWS.04Finding the engineering behind model latencyAn audit traced roughly 70% of a 50 to 60 second response time to prompt construction, orchestration and cold starts. Rebuilding the serving path reduced latency by 50% at equal task performance.05Governing agents acting on behalf of usersA self-hosted platform where effective access is the intersection of user and agent rights across C1-C4 data, with revocable delegation and auditable decisions deployed under GitOps on Kubernetes.
- 50%
- latency reduction at equal task performance
- 4
- AI use cases taken to production on EKS
- C1-C4
- classified data governed for delegated agent access
Next role
Building something difficult?
I am looking for a Solutions Architect or AI Infrastructure role inside a product team, with ownership that continues after the first deployment.