From Agents to APIs: Building Production-Ready AI Systems with Google ADK

Join GDG Sydney for a hands-on workshop exploring how to build practical AI systems with Google ADK and FastAPI. In this session, we’ll move from building a basic agent to exposing it as a production-ready API, with a focus on developer-friendly workflows, clean architecture, and real-world deployment patterns.

This workshop is designed for developers, AI builders, and technical enthusiasts who want to learn how to connect agent logic with modern Python web APIs. Whether you are experimenting with LLM agents or planning to ship AI-powered services, this session will give you a solid starting point.

About the Speaker
Ashmi Banerjee is an Applied AI researcher and doctoral candidate at the Technical University of Munich, where her work focuses on recommender systems, human-centered AI, and responsible machine learning. She brings both academic depth and practical industry experience to her work, with a strong interest in building AI systems that are trustworthy, useful, and grounded in real-world needs. She is also a Google Developer Expert in Machine Learning and a Women Techmakers Ambassador, reflecting her active leadership in the broader tech community.

About the Workshop
Agent-based AI systems help structure complex LLM-powered applications, but integrating them into real-world systems remains a practical challenge. In this workshop, participants build a travel-planning AI agent using Google ADK and Python, moving from interactive experiments to production-ready execution. We cover agent runtimes, async programming, and exposing agents via FastAPI. The workshop is hands-on and code-focused, aimed at Python developers who want to build, debug, and deploy AI systems.

What you will learn:

What an AI agent is and how Google ADK structures one.

How to set up a local development environment for the workshop.

How to manage conversation state and run an agent pipeline.

How to connect an agent to FastAPI endpoints.

How to run the FastAPI server locally and test the API.

How this setup supports deployment, containerization, and real-world AI applications.

Prerequisites

Don't forget your laptops!

Python 3.10 or later.

An IDE such as VS Code or PyCharm.

uv installed, if you don’t already have it.

A Google API key or Vertex AI credentials, depending on your setup.

Come along to learn, build, and connect with the GDG Sydney community while exploring a modern stack for AI agent development!

Disclaimer

A ticket does not guarantee entry; entry is first come, first served.

Bring your laptops; this is a hands-on event.

We're here to help: If you have physical, learning, or accessibility requirements, please reach out via <mranocchiari@google.com> or <gdgsydneyau@gmail.com>

Agenda

---

Speaker

Ashmi Banerjee - Technical University of Munich (Google Developer Expert (GDE) ML, Women Techmakers Ambassador)

Doctoral Candidate focusing on Recommender Systems & HCI at Technical University of Munich, Google Developer Expert in Machine Learning
Ashmi Banerjee is a doctoral researcher at the Technical University of Munich, where her research focuses on Recommender Systems and Human-Computer Interaction. She graduated with a master's degree in Computer Science in 2019 from the same university…

Host

Kunal Ostwal - GDG Sydney (Organiser)

Complete your event RSVP here: https://gdg.community.dev/events/details/google-gdg-sydney-presents-from-agents-to-apis-building-production-ready-ai-systems-with-google-adk/.