AI agent design patterns

Key Takeaways

Hi everyone, welcome to this agentic pattern [music] series. So if you building with AI, you probably wonder how to really design the agentic system. Sometimes you need a single agent and other times you need a whole team of them working together. And today we will diving into AI agent design patterns. And each pattern we're going to provide you with practical example of code. And we're also going to walk through a live demo. And by the end of this agendic pattern series, you will learn how to build agentic solution from single agent pattern to different multi- aent pattern. And for today's episode, we will focus on practical examples of single agent, sequential agent, and parallel agent. For next episode, we're gonna cover orchestrator pattern, review and critique pattern with loop agent and agent as tool. All right, let's dive in. So let's begin with the most fundamental pattern, the single agent. Imagine you want to plan a trip. So with a single agent, you want to give it instruction on how to use a tool and we can have a set of tool like how to check the weather, check the traffic and schedule. But for our example today, we will simplify things with only Google search tool. So you will write a comprehensive prompt telling the agent how to plan the trip using the search tool. So the agent then relies on the model's reasoning capability to figure out the sequence of steps. As you can see from the screen, the code for this is very straightforward with ADK agent development kit. And let's test it with ADK web UI by typing ADKVAB. So if I type plan a trip to San Francisco, we can see in the tracing that agent is using the search tool to gather all the necessary information in just one go. We can see how it is using the tool from the tracing tab over here. However, this works for single task. But as you have more tools, all the tasks are getting more complex. For example, if I have a request finding a sushi in San Francisco that's open late and finding the fattest way getting there. So if I build this with single agent, we need to define the logic in system instruction with this massive prom and the behavior can become very unreliable. Since AI is nondeterministic, you cannot always guarantee that you will follow your multistep logic perfectly every time. So the single agent lack of control if is its main weakness with single agent pattern. The benefit is it is very simple to implement and is great for straightforward multi-step tasks. However, it is less reliable for complex workflow and is harder to control and can fill as tasks becoming more complex. Now, let's get to our second part of this video which is the sequential agent. So, with the task we covered earlier, how do we adding more control? This bring us to our first motion pattern. the sequential agent and this pattern is for highly structured repeatable task because the order of the operation is fixed. So the output of one sub aent become the direct input for the next sub aent. It's like an assembly line and for our trip we can break down to two specialized agent. So we have this first a foot finding agent and the second a transportation agent. This sequential agent ensures that we always run this food finding agent first and then this transportation agent next. So this gives us predictable reliable execution. And you can take a look at the screen for this code example of how to write it in sequential agent. All right, let's try it out in ADK web UI. And you can look at the tracing tab. You can see that it's executed the food agent first and then the transportation agent next. Perfect. Uh so how do they communicate with each other? So they share information through this shared session state which act like a shared scratch pad. You can check the session state value at this tab. And once the first agent writes its finding and then the second agent reads it from it by using this curly braces in its system prompt. So this is a form of short-term memory for your agent system. And the advantages of sequential agent is that it has high degree of control and reliability. It is more predictable than a single agent, but it can be very inflexible. This rigid predefined structure can adapt to dynamic situations. Now let's go to the third part of this video, the parallel agent. So what if some tasks don't need to be happen in order? Let's say if I want to plan a full trip need to find a good museum, find a good concert and then a good restaurant. If I doing them in sequential order, it will be slow. This is where the parallel agent pattern really shines. It allows multiple specialized agent to run independently at the same time. We can have three agent museum finder, concert finder, restaurant finder, all searching concurrently. As you can see, this will be a lot faster compared to doing them in sequential order. Of course, after they all find something, we need to bring this results together. So, a common approach is combine this with a sequential agent. So first we will run the search in parallel and then second we will run a final aggregator agent to synthesize all the results in a single trip plan. And here's the agent code on how to put together them. All right, let's test ADK web UI. After we type the question from this tracing tab, we can see three searching agents kick off all at the same time. So once they're done, the results are returned to the session state as you can see from this tab and then the final summarizing agent reset state and generate our plan and this logic is defined in the system prompt over here. As you can see this is a great way to reduce latency for task that can be broken down into independent subtask. So the benefit of this design is very obvious that it significantly reduce latency by running tasks at the same time. However, it can have higher initial cost because it is running multiple agents all at once and in a lot of use case it requires gather or synthesize step to combine the result which can add complexity to our design. All right, let's recap what we covered so far. So we covered the single agent. It is simple to implement. It can be very flexible but it lacks certain control over this whole system. We also cover the sequential agent. It add a certain level of control and make the system more reliable but it is not very flexible. And lastly we covered the parallel agent. It is fast, efficient and great for independent task but it can add cost and complexity to your system. With those pattern you can already build some powerful workflow. And in our next video we're going to level up with more advanced pattern for handling even more complex and dynamic problem. We will explore cases with different practical examples and the pattern we're going to cover next episode including uh the loop and critique pattern for selfcorrection, the coordinator pattern for dynamic routing [music] and also the powerful concept of using agent as a tool. All right, I will see you in [music] next video. Bye. >> [music]