aks-demos

Hands-on Azure Kubernetes (AKS) demos for learning and technical presentations.

This repository is to share my projects presented in my blog, youtube videos and tech presentations.

• provision AKS and related Azure resources,
• run KAITO model workspaces,
• deploy and test RAGEngine,
• expose services via ingress,
• build practical chat/RAG demo workflows.

---

Repository table of contents

aks-kaito

Main end-to-end AKS + KAITO + RAG demo project.

• aks-kaito/README.md: Full runbook and demo flow.
• Infrastructure and setup scripts:
  • aks-kaito/1-setup-aks-kaito-v0.80.sh
  • aks-kaito/2-setup-nginx-ingress-controller.sh
  • aks-kaito/3-setup-kaito-rag.sh
  • aks-kaito/4-streamlitapp-demo.sh
  • aks-kaito/4-deploy-gptoss-inference.sh
  • aks-kaito/test-kaitochatdemo.sh
• Key manifests:
  • ingress-*.yaml (routing)
  • *-workspace*.yaml (workspace deployments)
  • bge-small-ragengine.yaml (RAG engine)

Nested subprojects in aks-kaito:

• aks-kaito/ragengine-ingest-docs

  • aks-kaito/ragengine-ingest-docs/README.md
  • CLI for document ingestion, index management, and RAG chat/query.
  • Includes sample docs in aks-kaito/ragengine-ingest-docs/docs.

• aks-kaito/streamlit-chatbot

  • aks-kaito/streamlit-chatbot/README.md
  • Streamlit UI for OpenAI-compatible chat + RAG diagnostics.
  • Multipage app components in aks-kaito/streamlit-chatbot/pages.

• aks-kaito/kaito-rag-app

  • Additional app workspace for KAITO/RAG experimentation.

aks-keyvault

AKS + Azure Key Vault integration demos.

• aks-keyvault/README.md: Scenario walkthrough and references.
• Scripts:
  • aks-keyvault/aks-keyvault.sh
  • aks-keyvault/aks-keyvault-secret-polling.sh

---

Suggested learning path

1. Start with aks-kaito/README.md for the full AKS + KAITO + RAG deployment flow.
2. Use aks-kaito/ragengine-ingest-docs to ingest test documents and run RAG queries.
3. Use aks-kaito/streamlit-chatbot for interactive demos and troubleshooting visibility.
4. Explore aks-keyvault to add secrets management patterns for production-like scenarios.

---

Notes

• This repo is intentionally demo-focused and optimized for teaching concepts.
• GPU-backed workloads can incur cloud cost; remember to clean up resources after labs/demos.