Suresh Michael
All workshops
Voice Agents

Building Voice Agents with Google Agent Development Kit

A DevFest code lab that builds a real-time tourist guide agent on Google ADK, from protocol choice through custom tools, interruption handling, and what it takes to scale it.

Delivered
January 14, 2026
Event
Google DevFest Sri Lanka 2025
Location
Sri Lanka
Format
Code lab
Audience
Developers new to real-time agents

The slide deck is locked

Sign in below to open the deck and the rest of the write-up.

Slides

What we covered

The session is built around one artefact: a Sri Lanka tourist guide agent that answers by voice, calls tools mid-conversation, and can be cut off mid-sentence without losing the thread. Everyone leaves having built it.

  • How AI communication moved from request-response to a persistent, interruptible stream.
  • The model ecosystem underneath: Gemini's native audio, and what ADK adds on top of it.
  • WebSockets versus WebRTC, and the cases where each is the right transport.
  • ADK's architecture: the high-level system, then the tool-calling loop that does the real work.
  • Writing it: initialisation, the WebSocket endpoint, custom tool definitions, and interruption handling.
  • Observability with LangFuse, so a non-deterministic conversation is still debuggable after the fact.
  • Deployment and scaling: what changes between one caller and a thousand.

Members only

Keep reading

The slide deck and the rest of this write-up are free, sign in with Google and they stay unlocked on this device.

No newsletter, no spam. Your email is used to keep you signed in.

Share this workshop
Copied