Agentic AI Engineering Lab #6 – CI/CD & Coordinating Agents in the Cloud

In our last sprint, we set up a shared baseline repository featuring a functional "Welcome Agent." We successfully used local shell scripts to build the Docker image, deploy it to GCP, and query the live endpoint. It's a solid foundation, but manually executing deployment scripts from a local machine is a bottleneck for rapid, collaborative development.

The Plan: This session is about automating our existing cloud deployment. We will take the exact build and deploy steps we currently run locally and translate them into a GitHub Actions workflow. Once our agents are updating automatically on every push to `main`, we will shift focus back to agent behavior. We will build a simple Orchestrator Agent that uses the A2A protocol to dynamically delegate tasks to our cloud-hosted workers.

This is a "Lab" session. We will:

• Automate the Pipeline: Convert our local Docker and `gcloud` deployment scripts into a seamless GitHub Action workflow.
• Build the Orchestrator: Develop a lightweight central agent designed to interpret a user prompt and route sub-tasks to our deployed workers.
• Live Multi-Agent Communication: Establish an active dialogue over the A2A protocol between your Orchestrator and your worker agents.

The Format: We run our sessions like an anthology series: Every meetup is a standalone sprint. While we often build on previous results, each session provides a complete experience and remains fully accessible to first-time attendees. This is a high-density micro-meetup limited to 6 seats. (Newcomers: We will provide the fully prepped template repository so you can jump straight into the automation phase!)

What you'll build By the end of the session, you'll have:

1. A Repeatable CI/CD Pipeline: A set-and-forget GitHub Action that handles the GCP deployment automatically on every commit, replacing the need for local scripts.
2. An Orchestrator Agent: A coordinator capable of breaking down requests and utilizing external agent tools via A2A.
3. A Live Multi-Agent Setup: A functioning environment where your Orchestrator successfully communicates with and aggregates results from your automated cloud workers.

Schedule (Saturday, 10:00 - 13:00 CEST)

• 10:00: Quick round of introductions.
• 10:15: Intro session: Migrating local scripts to GitHub Actions & Orchestrator routing patterns.
• 10:30: Build time: CI/CD automation & building the Orchestrator.
• 12:00: Live demos + technical teardown.
• 12:45: Wrap-up + takeaways.

Prerequisites

• Accounts: A Google Account (with GCP access*) and a GitHub Account (for Actions).
• Laptop with dev setup: Docker and Git installed and working.
• Access to an LLM API: (Gemini, Claude, OpenAI, etc.).
• Mindset: This is a lab for engineers, not a tutorial. We learn by building, breaking, and iterating.

Logistics

• Location: We meet at my office in the Grunewald area near Martin-Luther-Krankenhaus. The exact address will be shared privately with attendees to maintain privacy.
• Capacity: Limited to 6 seats.

* You currently get $300 in free credits when you sign up with GCP, which is plenty for what we're trying to do. We can give some initial pointers for how to get started during the meeting, but won't have time to do a full GCP setup with you.