AI Governance Lab is an open-source architecture and engineering lab dedicated to building governed, compliant, and scalable Enterprise AI systems across hybrid and multi-cloud environments.
📌 Public Organization Profile: Rendered from
profile/README.mdat https://github.com/AI-Governance-Lab.
- Sovereign AI Hubs & LiteLLM Gateways: Self-hosted Open WebUI, LiteLLM Proxy (
:4000), Apache Tika OCR, cloud-offloaded RAG, and domain microservices (private-enterprise-ai-hub). - High-Performance GPU Inference: NVIDIA NIM, TensorRT-LLM, Triton, vLLM, and Ollama across GCP GPU VMs and K3s/Kubernetes (
nvidia-nim-enterprise-patterns,k8s-ollama-fastapi-demo). - Agentic Infrastructure Automation & Multi-Cloud RAG: Qdrant vector search + guarded Terraform/Ansible execution on AWS EKS, Rancher, and IBM Red Hat OpenShift / PowerVS for SAP S/4HANA (
k8s-ai-agent-multicloud-rag-iac,s4hana-ai-agent-openshift,k8s-ai-agents-demo). - Quantum-Classical Hybrid Orchestration: Natural-language-to-circuit compilation (Google Cirq / IBM Qiskit) governed by an independent LLM-as-a-Judge (
governed-quantum-agents). - Google Cloud / Vertex AI Patterns: Agent Builder, Google ADK, Kubeflow/Vertex AI Pipelines, and serverless Cloud Run inference (
vertex-ai-agentbuilder-demo,vertex-ai-pipeline-demo,vertex-ai-cloudrun-demo). - Enterprise AI Evaluation, MLOps & Compliance: Continuous drift monitoring with Evidently AI and experiment/prompt lineage with MLflow 3.x mapped to EU AI Act, NIST AI RMF, ISO/IEC 42001, GDPR, and SOC2 (
k8s-evidently-demo,k8s-mlfow-demo,k8s-mlflow-rancher-desktop).
| Repository | Domain | Availability |
|---|---|---|
private-enterprise-ai-hub |
Self-Hosted Enterprise AI Hub, LiteLLM Gateway, Tika OCR & CAD Microservice | 🔒 Private (Public Release Soon) |
nvidia-nim-enterprise-patterns |
NVIDIA NIM & TensorRT-LLM Inference Patterns (GCP GPU VMs & K3s/Helm) | 🔒 Private (Public Release Soon) |
governed-quantum-agents |
Venus: Quantum-Classical Hybrid Orchestrator + LLM-as-a-Judge | 🌐 Public |
k8s-ai-agent-multicloud-rag-iac |
Multi-Cloud RAG + Terraform Runner Agent (Rancher & AWS EKS) | 🌐 Public |
s4hana-ai-agent-openshift |
Zero-Touch SAP S/4HANA Provisioning Agent (IBM ROKS, watsonx.ai, PowerVS) | 🌐 Public |
k8s-ai-agents-demo |
Reference Kubernetes AI Agent Services (Qdrant, FastAPI, MLflow) | 🌐 Public |
vertex-ai-agentbuilder-demo |
Google Vertex AI Agent Builder & Agent Development Kit (ADK) | 🌐 Public |
vertex-ai-pipeline-demo |
Vertex AI Pipelines (Kubeflow v2) MLOps Orchestration | 🌐 Public |
vertex-ai-cloudrun-demo |
Serverless Vertex AI Embeddings & Gemini Microservices on Cloud Run | 🌐 Public |
k8s-evidently-demo |
Evidently AI Drift & Quality Monitoring on Kubernetes | 🌐 Public |
k8s-mlfow-demo |
Production MLflow 3.x Cluster Stack (PostgreSQL + MinIO) | 🌐 Public |
k8s-mlflow-rancher-desktop |
Local Developer MLflow 3.x Stack on Rancher Desktop (k3s) | 🌐 Public |
k8s-ollama-fastapi-demo |
Lightweight Local LLM Serving with Ollama & FastAPI on Kubernetes | 🌐 Public |
Default license is Apache License 2.0 (see LICENSE and individual repositories).
Marian Ropota — Hands-On AI/ML & Cloud Platform Architect
🔗 LinkedIn • 🏅 Credly Certifications • 🐙 AI Governance Lab