Skip to content
View ppiova's full-sized avatar
🏠
Working from home
🏠
Working from home
89 followers · 103 following

Achievements

Achievement: Pair Extraordinairex4Achievement: Pull Sharkx3Achievement: QuickdrawAchievement: YOLOAchievement: Arctic Code Vault Contributor

Achievements

Achievement: Pair Extraordinairex4Achievement: Pull Sharkx3Achievement: QuickdrawAchievement: YOLOAchievement: Arctic Code Vault Contributor

Highlights

  • Pro

Block or report ppiova

Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
ppiova/README.md
header

Microsoft MVP in AI · 🐳 Docker Captain · Director of AI · Community Leader · Author

Shipping AI agents, enterprise GenAI and signed, multi-arch, SBOM-attested container images — built on Azure AI, Microsoft Foundry, MCP and Docker.

Microsoft MVP Docker Captain LinkedIn X / Twitter dev.to Meetup Book GHCR

Mission. Make enterprise-grade AI samples that teams can actually trust: reproducible, auditable, signed. No "works on my laptop" — everything ships as a multi-arch OCI image with SBOM and SLSA provenance, so security reviews become a one-liner.

⚡ Try it — and verify it — in 30 seconds

# Run a signed, multi-arch image (Blazor chat UI from the Docker Model Runner lab)
docker run --rm -p 8080:8080 ghcr.io/ppiova/docker-model-runner-lab/blazor-chat:latest
# then open http://localhost:8080

Every image is multi-arch, SBOM-attested and carries keyless-signed SLSA provenance — verify any of them end-to-end:

# Multi-arch manifest + attached SBOM & SLSA provenance
docker buildx imagetools inspect ghcr.io/ppiova/docker-model-runner-lab/blazor-chat:latest

# Verify the keyless-signed build provenance (GitHub OIDC, via Sigstore)
gh attestation verify oci://ghcr.io/ppiova/docker-model-runner-lab/blazor-chat:latest --owner ppiova

🚢 Shipping in the open

12 featured repos · 1 app on the Microsoft Store · 6 signed images on GHCR · SBOM + SLSA provenance · multi-arch (amd64 / arm64)

GHCR SBOM Provenance Multi-arch Signed CI

🏆 Recognition
🏅 Microsoft MVP in AI Public profile — MVP since 2022
🐳 Docker Captain Docker profile — recognized for expertise in containers & secure software supply chain
📘 Author AI-102 Certification Guide — Amazon
🎤 Community Leader Organizer at .NET Baires, one of LATAM's largest .NET & AI communities — co-organizer of AgentCon Córdoba, VS Code Dev Days and Global Azure Latino
💼 Director of AI OZ Digital Consulting — leading the AI Center of Excellence and enterprise GenAI adoption

🛠️ What I Build

  • AI agents, orchestration and multi-agent patterns
  • Enterprise Generative AI on Azure AI and Microsoft Foundry
  • MCP servers and developer workflows
  • Containerized AI tooling — devcontainers, compose stacks, signed OCI images
  • Secure software supply chain for AI: SBOMs, SLSA provenance, reproducible CI
  • Reference implementations that scale from demo to production

🧰 Tech Stack

.NET C# Python Azure Azure OpenAI Docker GitHub Actions MCP

📌 Featured Work

🔹 mcp-servers-microsoft-ecosystem  stars Community-curated catalog of MCP servers across the Microsoft ecosystem (Azure, M365, Fabric, Power Platform, GitHub, Copilot Studio) — complements the official microsoft/mcp with community implementations, clients, starters and security guidance.

🔹 AgentFrameworkDemos  stars AI agents, orchestration and real-world patterns with Microsoft Agent Framework on .NET.

🔹 AgentFX-MCP-MicrosoftLearn  stars Reference integration between Agent Framework and the Microsoft Learn MCP Server.

🔹 TravelMCP  stars End-to-end MCP server illustrating real-world tool workflows.

🔹 AzureOpenAI-EntraID-ConsoleApp  stars Enterprise-grade Azure OpenAI auth with Entra ID — production-ready patterns.

🔹 AI-Custom-Avatar  stars Azure AI Avatar Agent on .NET 10 + Blazor + .NET Aspire — Speech-to-Text, Azure OpenAI and a talking avatar with synchronized lip-sync. Built with Bruno Capuano.

🪟 Published on the Microsoft Store

🔹 VMAzureApp — Cloud VM Manager for Windows  stars  Microsoft Store Manage Azure VMs across all your subscriptions without opening the portal — start / stop / restart / hibernate with bulk actions, local scheduling and real-time power-state monitoring. WPF on .NET 10 with Entra ID sign-in, shipped as MSIX through GitHub Actions. Get it from the Microsoft Store.

🐳 Docker-first series — Microsoft Agent Framework

🔹 agent-framework-devcontainer  stars Single-agent starter. Multi-stage Dockerfile (Alpine, non-root), Dev Container / Codespaces ready, docker compose up and you're running.

🔹 mcp-docker-starter  stars Polyglot compose: Python FastMCP server + .NET Agent Framework client, connected via SSE over a private bridge network. Service discovery, healthchecks, non-root everywhere.

🔹 ai-agents-compose-stack  stars Multi-agent workflow (Researcher → Writer) with Microsoft Agent Framework, OpenTelemetry + Aspire Dashboard observability — fully in Docker Compose. Published image ships multi-arch with SBOM + SLSA provenance.

🐳 Docker Model Runner — run LLMs locally

🔹 docker-model-runner-lab  stars Hands-on lab for Docker Model Runner — run LLMs locally behind an OpenAI-compatible API with .NET and Docker Compose. No cloud, no API keys, nothing leaves your machine.

🔹 docker-model-runner-multi-sdk-demo  stars One endpoint, three SDKs: DMR serves the same local model through OpenAI, Anthropic and Ollama formats on a single port — dev/prod parity with Microsoft Foundry, six runnable Python + .NET examples.

🔐 Container Supply Chain

🐳 As a Docker Captain, I focus on making containers reproducible, portable and trustworthy for real AI workloads — the practices below are exactly what that recognition is about.

Every image listed below ships from GitHub Actions with:

Guarantee How
Multi-arch docker buildx — linux/amd64 + linux/arm64
SBOM SPDX attestation attached (buildx sbom: true)
Provenance SLSA build provenance attached (buildx provenance: true)
Signed Keyless build-provenance attestation via GitHub OIDC — Sigstore (actions/attest-build-provenance)
Reproducible Pinned bases, deterministic builds in GitHub Actions

Published images

# Image Source
1 ghcr.io/ppiova/agent-framework-devcontainer agent-framework-devcontainer
2 ghcr.io/ppiova/mcp-docker-starter/mcp-server mcp-docker-starter
3 ghcr.io/ppiova/mcp-docker-starter/agent-client mcp-docker-starter
4 ghcr.io/ppiova/ai-agents-compose-stack ai-agents-compose-stack
5 ghcr.io/ppiova/docker-model-runner-lab/compose-api docker-model-runner-lab
6 ghcr.io/ppiova/docker-model-runner-lab/blazor-chat docker-model-runner-lab
🔎 Inspect any image 
# Manifest + attestations
docker buildx imagetools inspect ghcr.io/ppiova/<image>:latest --format '{{ json . }}'

# SBOM
docker buildx imagetools inspect ghcr.io/ppiova/<image>:latest \
  --format '{{ json .SBOM }}'

# Provenance
docker buildx imagetools inspect ghcr.io/ppiova/<image>:latest \
  --format '{{ json .Provenance }}'

# Verify the keyless-signed build provenance (GitHub OIDC, via Sigstore)
gh attestation verify oci://ghcr.io/ppiova/<image>:latest --owner ppiova

Why this matters. Enterprises can't deploy unsigned samples. Shipping every image with SBOM + provenance turns "interesting demo" into "something I can put a ticket on."

🎤 Speaking & Teaching

Regular speaker at Microsoft Reactor, Microsoft Ignite, Microsoft Build and community events across LATAM. Highlights:

📚 More sessions & community events (2023–2026)

Microsoft Reactor

Agentes & AI avanzada

Fundamentos de IA Generativa

IA + .NET — Reactor LATAM series

Dev & AI tooling

Series & panels

Microsoft Ignite & Build

Community & Conferences

✍️ Writing & Content

📘 AI-102 Certification Guide — practical guide for building real-world solutions with Azure AI, Generative AI and Microsoft Foundry.

📝 Recent articles:

Full technical series on dev.to and LinkedIn — covering Azure AI, Agents, MCP, Foundry and containerized AI.

🤝 Open Source Contributions

Contributor to Microsoft and community OSS projects around AI and Agents, including:

🎓 Workshops & Engagements

Available for:

  • Hands-on workshops — Azure AI Foundry · Agent Framework · MCP · Docker supply chain for AI
  • Conference talks & keynotes — GenAI, Agents, secure container delivery
  • Advisory — enterprise GenAI adoption, AI Center of Excellence setup

📫 Reach out: LinkedIn · X / Twitter

Focused on making AI practical, verifiable and usable for real-world applications.

Pinned Loading

  1. mcp-docker-starter mcp-docker-starter Public

    Containerized MCP server (Python/FastMCP) + Microsoft Agent Framework .NET client orchestrated with docker-compose.

    Python 1

  2. agent-framework-devcontainer agent-framework-devcontainer Public

    Docker-first starter for Microsoft Agent Framework: multi-stage Dockerfile, devcontainer, Codespaces-ready single-agent sample with Azure OpenAI.

    C# 1

  3. AgentFrameworkDemos AgentFrameworkDemos Public

    AgentFrameworkDemos (.NET + Microsoft Agent Framework)

    C# 3

  4. Microsoft-Azure-TTS-Avatar Microsoft-Azure-TTS-Avatar Public

    JavaScript 1

  5. VMAzureApp VMAzureApp Public

    Windows desktop app to manage Azure virtual machines

    C# 1

  6. docker-model-runner-lab docker-model-runner-lab Public

    Hands-on lab for Docker Model Runner: run LLMs locally with an OpenAI-compatible API, .NET and Docker Compose

    CSS 4