Mood: Let‘s build a record-breaking cloud-native AI supercomputer from scratch.I'm all in for Dev & MLOps + connected infrastructure.Kubernetes enthusiast since 2018, switched from TensorFlow to PyTorch in 2020. Let's get this whole AGI thing going (wherever it leads..).Always on the hunt for new architectures/frameworks/processes.I love to work in interdisciplinary teams of Dev/MLOps Engineers and Data Scientists. My focus lays on building the technical multi-cloud infrastructure for BI apps & data scientist in the context of AI data & compute infrastructure (kubernetes based HPC in the context of microservice architectures). I can get you from a bare metal kubernetes cluster (kubeadm) for BI apps up to large scale multi-cloud GPU clusters integrated with an experiment management system (Polyaxon), distributed training & deployment (KubeFlow) and reproducible data management & versioning (Pachyderm).Skills- Infra/DevOpsKubernetes, Go, Microservice Architectures (Kubeadm, Rancher, Rook/Ceph, SeaweedFS, CRI-O, Ingress, Loft/vCluster, Prometheus/Grafana, Harbor, ArgoCD, Jaeger, Keycloak, Velero, nVidia GPU Container Driver, MinIO, Verdaccio, Dex, ..)- AI/MLOpsPyTorch (Python), Polyaxon, KubeFlow, Pachyderm, Git based CI/CD architectures for ML- BasicsLinux server/network administration (Ubuntu), Bash, GitLab, Jira, Confluence, SCRUM, OpenVPN, [missing little things]PS: I am a huge fan of Transformers and had an account at OpenAI before I think it was popular.