Introducing Agent Development Kit

Hey agent, can you help me plan my trip to San Francisco? Okay, I can help with that. What's your budget for this trip?

that. What's your budget for this trip?

Okay, let me pause it there and tell you that I just interacted with a multimodal multi-agent application authored in under 100 lines of code with a new agent development kit. Let's get into

development kit. Let's get into [Music] it. Hi, I'm Anander, a product manager

it. Hi, I'm Anander, a product manager here at Google Cloud, and I'm really excited to share something new with you today. We talked to a lot of AI

today. We talked to a lot of AI developers and hear consistently that building sophisticated agents particularly multi-agent or multimmodal ones, is often way more complex and

timeconuming than it should be. We felt

that friction firsthand building agents internally for our customers. And that's

exactly why today we're thrilled to release a brand new open-source project the agent development kit or ADK for short. It is built directly from

short. It is built directly from Google's experience creating production grade agents. Our goal with ADK is

grade agents. Our goal with ADK is simple. To give developers a powerful

simple. To give developers a powerful open foundation for building complex productionready AI agents intuitively

and with greater control. To us a truly open foundation means it is model agnostic. It's deployment agnostic and

agnostic. It's deployment agnostic and built for interoperability.

So starting with model agnostic, look we obviously love our Gemini models and especially the latest Gemini 2.5 Pro model which has amazing reasoning

capabilities. But with ADK, you can

capabilities. But with ADK, you can choose any model you want to use whether it's a model from Google or anywhere else. ADK is also deployment agnostic.

else. ADK is also deployment agnostic.

It can run locally wherever you do development or on Google cloud or wherever your infra happens to live. It

is also built for interoperability. So you can easily

interoperability. So you can easily integrate your agents with your existing tools and services or even integrate with agents that you may have built on

other frameworks. Now Bo, you led the

other frameworks. Now Bo, you led the engineering effort here. Why don't you share some of the design thinking behind ADK? Sure. Thanks Annette. Hi, I'm Bo

ADK? Sure. Thanks Annette. Hi, I'm Bo Yang, tech lead for agent development kit. One of our core principles behind

kit. One of our core principles behind this framework was to make agent development feel like software development. We deliberately built ADK

development. We deliberately built ADK using familiar software engineering patterns just like regular classes or functions. Can I just say thank you?

functions. Can I just say thank you?

Seriously, Bo, as a developer that wants to build a gentic application for production, that's just great to hear.

Of course. Also looking ahead, we saw huge potential for agents that interact naturally. Agent that can hear, can see

naturally. Agent that can hear, can see can respond in real time. That's why

native birectional audio and video streaming is fundamental and it's baked into the core SDK. This unlocks ability to create

SDK. This unlocks ability to create truly interactive and real-time agent experiences. Another common friction

experiences. Another common friction that we often hear is that it is require a lot of setup to just test and debug

agent locally. Developers wanted to

agent locally. Developers wanted to iterate quickly without spinning up a ton of complex dependencies. So we included a UI

dependencies. So we included a UI playground right in SDK. Now with a single command ad web, you can launch a

UI on your laptop to test, visualize and debug your agent. By the way, you can be up and running in seconds

focusing purely on your agent's technology. Agent Development Kit is

technology. Agent Development Kit is available today as a Python SDK with more languages like Java coming soon.

With this ADK, you can build sophisticated multi- aent applications in just under 100 lines of code. Okay

Ivan, do you want to show us a demo?

Absolutely. So, let's do this. So, Anan

and Bo talked about simplifying complexity. So, let's build a trip

complexity. So, let's build a trip planner agent super quickly. It's a

multi-agent setup. We will have a main trip planner agent orchestrating two sub agents. An idea agent to brainstorm a

agents. An idea agent to brainstorm a fun trip destination and a reefine agent to check if those idea fits with a given

budget. Okay, so let's dive into the

budget. Okay, so let's dive into the code. Here I have my two uh agent

code. Here I have my two uh agent definitions and it's just standard ADK agent classes. So we give them an LLM

agent classes. So we give them an LLM some instructions and more importantly clear description. So the parent agent

clear description. So the parent agent understand their capabilities. So now we have our definition. They look good.

They're super concise. So let's learn it locally. Like boom mentioned, it's just

locally. Like boom mentioned, it's just one command in the terminal adk web. So

let's try it. And boom. Okay. Howard local dev uh

it. And boom. Okay. Howard local dev uh UI is now ready. So let's give our planner agent a

task. Hello. So just a fun

task. Hello. So just a fun weekend trip for from San Francisco under

$200. Let's see. But it's a tricky It's

$200. Let's see. But it's a tricky It's a tricky question. Challenging. Yeah

it's going to be challenge. Let's

see. And there we go. We got a response.

So now we have our agent working. And if

you're a developer, you want to know what's happening under the hood. The

event tab shows the agents workflow and how it decided to use the sub agent.

Here you can see that the parent agent delegated to gather ideas and do a budget check. The request and response

budget check. The request and response tab give us the stepbystep trace so you know what data goes where. Yes, that's

super useful like having this tab that allows me to know what the action agent takes, what kind of function calls he had. Uh it's very it's very useful and

had. Uh it's very it's very useful and it's actually essential for debugging especially with multi- aents uh when you have multi- aents in interacting between

each other. So having this level of

each other. So having this level of visibilities locally is just fantastic.

So now uh Bo also mentioned that uh we have these features called niva birectional audio and video streaming which essentially means that agent can hear and see your input. It's built in

no extra libraries needed for the basic setup. So let's try the audio. Anan why

setup. So let's try the audio. Anan why

don't you ask a question? Absolutely.

Yeah. Let's uh let's take it for a spin.

Can you suggest an itinerary for two days in Seattle?

I can definitely help you plan a two-day trip to Seattle. What kind of activities do you enjoy? Yeah, what's impressive about this is how quickly you spun up this multi- aent application to have a

live audio conversation. Um, and I don't think you had to write any special code to get all this audio streaming working.

No, that's right. ADK handles all the underlying streaming mechanism for me behind the scene. And as a developer this speed up uh massively my development. it it let me just focus on

development. it it let me just focus on the agent logic which in this case is planning the trip and the core agent definition is still only 30 line of

codes for those three agents combined we define them we spin up the the bug UI and we had a live audio conversation all running locally and that's just a

glimpse of what's possible with agent development kit you have the simplicity control flexibility and that multimodel streaming capability native to the SDK.

And what we've built with Agent Development Kit goes beyond just an SDK.

It's it's a robust open foundation built directly on Google's production experience. We're investing in the

experience. We're investing in the ecosystem to empower developers like you to build complex powerful agentic solutions. So if you want to dive in, we

solutions. So if you want to dive in, we got the links right below in the description and check out the ADK documentation for guides and concepts.

Or if you're just like me and you prefer to jump straight into the code, head over to the GitHub repo. We included

plenty of sample agents there to help you to get started with it. Great. Thank

you for tuning in and we're excited to see what you'll build with the agent development kit.

[Applause] [Music]