1 00:00:00,400 --> 00:00:03,840 Okay, so now we get to the meat of Autogen core. 2 00:00:04,240 --> 00:00:11,360 We are going to create a runtime, a single threaded agent runtime it's called, which is a standalone 3 00:00:11,360 --> 00:00:17,640 runtime running on my computer, which as it says, will handle agents in a single threaded way. 4 00:00:18,160 --> 00:00:22,640 And the first thing that we're going to do is register. 5 00:00:22,640 --> 00:00:28,920 We call register on the agent itself to say, I want you to register yourself with this runtime. 6 00:00:29,040 --> 00:00:31,560 Now, this isn't creating an agent. 7 00:00:31,600 --> 00:00:34,440 This is just saying you are a type of agent. 8 00:00:34,440 --> 00:00:40,680 And I want you to tell the runtime that you are a type of agent that can be spawned, you can be created, 9 00:00:40,920 --> 00:00:45,560 and you are a type of agent of type simple agent that's going to be your type. 10 00:00:45,920 --> 00:00:52,920 And this, this thing here, this is a function which can generate new versions of you. 11 00:00:52,920 --> 00:00:54,240 It can instantiate you. 12 00:00:54,280 --> 00:00:55,360 It is a factory. 13 00:00:55,640 --> 00:00:57,600 And you pass that in as well. 14 00:00:57,840 --> 00:00:58,960 And so I will do this. 15 00:00:59,000 --> 00:01:03,660 It's now created a runtime and it's created a type of agent called a simple agent. 16 00:01:03,700 --> 00:01:05,980 We haven't actually yet built any of these. 17 00:01:05,980 --> 00:01:07,300 We haven't instantiated one. 18 00:01:07,300 --> 00:01:12,660 But the as a type of agent is now a known thing to our runtime. 19 00:01:12,700 --> 00:01:15,500 Our runtime knows that there are such things as simple agents. 20 00:01:15,740 --> 00:01:18,380 And we're now going to start our runtime. 21 00:01:18,380 --> 00:01:20,220 It is now running. 22 00:01:20,220 --> 00:01:21,980 It is a it is a proper runtime. 23 00:01:22,460 --> 00:01:31,460 And now we are going to uh, we're going to first of all create an agent ID object to identify an agent. 24 00:01:32,100 --> 00:01:34,340 And we'll we'll do this properly. 25 00:01:34,380 --> 00:01:41,100 We'll say agent ID equals that, that is an ID that would identify this. 26 00:01:41,100 --> 00:01:45,980 And we are going to say that we want the default agent. 27 00:01:47,100 --> 00:01:52,860 Uh, because, because that, that will then, uh, we'll make sure that we have an agent created, 28 00:01:53,140 --> 00:01:59,860 and then we're going to send a message to, we're going to send a message to the, to the agent, which 29 00:01:59,860 --> 00:02:04,630 is called simple agent, The default ID and we're going to send a message. 30 00:02:04,630 --> 00:02:06,550 Well hi there to it. 31 00:02:06,750 --> 00:02:09,990 And then we will print whatever comes back. 32 00:02:10,470 --> 00:02:12,790 So any ideas what's going to come back I wonder. 33 00:02:14,590 --> 00:02:18,950 Uh, so uh it replied this is simple agent default. 34 00:02:18,950 --> 00:02:24,830 So it's simple agent is its type, which is the same one we registered and it default is the, uh, 35 00:02:24,990 --> 00:02:29,150 the, the id uh, and it said, you said, well hi there. 36 00:02:29,190 --> 00:02:30,310 And I disagree. 37 00:02:31,510 --> 00:02:32,190 Charming. 38 00:02:32,630 --> 00:02:34,990 Uh, and that is uh, that's it. 39 00:02:34,990 --> 00:02:39,870 So there is a demo of auto gen core and that's what auto gen core does. 40 00:02:39,910 --> 00:02:44,950 And putting in agent functionality like having code in there that calls LMS. 41 00:02:44,990 --> 00:02:46,510 That's your job or my job. 42 00:02:46,510 --> 00:02:48,950 That's not auto gen core's job. 43 00:02:49,070 --> 00:02:54,630 What it's there to do is to handle the passing of messages around by looking things up based on their 44 00:02:54,630 --> 00:02:57,230 type and their ID, and that's what it does. 45 00:02:57,510 --> 00:03:03,770 So now I'll show you one which has an LM behind it, but it shouldn't be any surprise to you because 46 00:03:03,770 --> 00:03:05,370 it's just the same infrastructure. 47 00:03:05,370 --> 00:03:13,370 The fact that we're calling an LLM is is almost of no consequence to, to, uh, the Autogen core framework. 48 00:03:13,530 --> 00:03:19,050 So what we're going to do is we're going to make a slightly more interesting one called my LM agent, 49 00:03:19,090 --> 00:03:21,210 still a subclass of rooted agent. 50 00:03:21,450 --> 00:03:24,050 And we're going to use a proper LM. 51 00:03:24,050 --> 00:03:30,210 And we could just call OpenAI directly using OpenAI dot create. 52 00:03:30,250 --> 00:03:32,490 We could just call the Python client library. 53 00:03:32,690 --> 00:03:39,490 But we can also use Autogen agent chat and we might as well since it's autogen week. 54 00:03:39,810 --> 00:03:46,090 So this is the same kind of thing, a subclass of rooted agent that's going to be two functions again 55 00:03:46,130 --> 00:03:49,330 init and handle my message. 56 00:03:49,570 --> 00:03:53,450 So the init just just calls the superclass. 57 00:03:53,770 --> 00:03:56,610 But it then creates a model client. 58 00:03:56,890 --> 00:03:58,050 But this could be doing anything. 59 00:03:58,050 --> 00:04:00,650 But this is the Autogen agent chat way. 60 00:04:00,690 --> 00:04:05,070 We're going to create one of these things, a GPT four mini model client. 61 00:04:05,070 --> 00:04:10,190 And we're going to set a field called underscore delegate, which we're just sort of holding on to. 62 00:04:10,230 --> 00:04:12,470 This is this is the the underlying. 63 00:04:12,510 --> 00:04:18,390 This is what our agent object is going to delegate to when it actually needs some code, some, some 64 00:04:18,430 --> 00:04:19,550 an LLM to run. 65 00:04:19,550 --> 00:04:21,110 And we could have anything we want. 66 00:04:21,110 --> 00:04:26,670 We could, we could do we could reply back with random things or we could create an instance of assistant 67 00:04:26,670 --> 00:04:27,030 agent. 68 00:04:27,030 --> 00:04:28,310 And that's what we're going to do. 69 00:04:28,550 --> 00:04:29,870 And that's where we put in. 70 00:04:29,910 --> 00:04:33,750 If you remember, when you create an assistant agent, there's the name of the agent and there's the 71 00:04:33,750 --> 00:04:35,350 model client, and there it is. 72 00:04:35,550 --> 00:04:43,190 And as a result in underscore delegate underscore often used to show sort of like private secret, uh, 73 00:04:43,510 --> 00:04:47,230 variables that other people should know about, like, like private in the, in the Java world. 74 00:04:47,470 --> 00:04:52,110 Um, so the this is being set to an assistant agent. 75 00:04:53,270 --> 00:04:57,790 So, uh, now we just have to write our message handler. 76 00:04:58,070 --> 00:05:00,350 And as I say, it can be called anything you like. 77 00:05:00,510 --> 00:05:02,640 I'm calling it handle my message type. 78 00:05:03,040 --> 00:05:08,680 Um, and that's really to sort of draw your attention to the fact that what matters is what type of 79 00:05:08,680 --> 00:05:16,800 object comes in the message field, because Autogen core will automatically route messages to agents 80 00:05:16,800 --> 00:05:24,680 that should receive them based on the the finding a handler that that looks for a message of that type. 81 00:05:24,840 --> 00:05:25,520 That's how it works. 82 00:05:25,520 --> 00:05:27,320 That's the clever routing that it does. 83 00:05:27,360 --> 00:05:31,200 And by the way, you can send a message direct to an agent, which is what we're doing right now. 84 00:05:31,200 --> 00:05:37,120 But it also has a whole kind of pub sub thing there where you can subscribe to topics and you can publish 85 00:05:37,120 --> 00:05:41,200 out messages to lots of agents that are all interested in one particular topic. 86 00:05:41,360 --> 00:05:48,160 So the whole process of dispatching messages to the right agent and then calling the right handler, 87 00:05:48,320 --> 00:05:49,560 that's the clever stuff. 88 00:05:49,560 --> 00:05:50,960 That's what it does well. 89 00:05:51,000 --> 00:05:55,120 And why why we use a framework like this rather than coding it ourselves. 90 00:05:55,360 --> 00:05:57,520 But for some of these examples, it's pretty trivial. 91 00:05:57,520 --> 00:05:59,320 So we could equally code it ourselves. 92 00:05:59,640 --> 00:06:03,820 But you can imagine that this is done in a way that can be scaled and distributed. 93 00:06:04,100 --> 00:06:06,740 So anyways, handle my message type. 94 00:06:07,100 --> 00:06:14,100 Uh, as long as, uh, message is sent to this agent with the ID and and type, and that the object 95 00:06:14,100 --> 00:06:17,580 that's sent is of this type, it will arrive at this function. 96 00:06:17,780 --> 00:06:21,820 And I'm going to say that I received a message and print the content. 97 00:06:22,340 --> 00:06:24,700 Then I'm going to now now this is confusing. 98 00:06:25,300 --> 00:06:31,020 This object here text message looks subtly different to this object here. 99 00:06:31,020 --> 00:06:31,780 Message. 100 00:06:32,180 --> 00:06:34,300 Uh, do you know what's going on here? 101 00:06:34,500 --> 00:06:39,180 This text message is text message from the, um, autogen. 102 00:06:39,500 --> 00:06:40,460 It's this thing here. 103 00:06:40,460 --> 00:06:42,060 It's the agent chat messages. 104 00:06:42,060 --> 00:06:43,020 The text message. 105 00:06:43,020 --> 00:06:50,020 The same thing we were using yesterday and the day before when we were interacting with agent chat agents. 106 00:06:50,020 --> 00:06:55,260 So this is a autogen agent chat, uh, message, which can be a bit confusing. 107 00:06:55,260 --> 00:06:56,020 And you might. 108 00:06:56,220 --> 00:06:59,900 Yeah, there are ways you could you could get around that by, by giving it a different you can say 109 00:06:59,900 --> 00:07:08,960 like as agent chat text message if you wanted to to like make it so the code is a little bit clearer 110 00:07:08,960 --> 00:07:13,560 and you're distinguishing between the Autogen core and the Autogen agent chat. 111 00:07:13,880 --> 00:07:14,520 All right. 112 00:07:14,960 --> 00:07:16,080 Anyways, so sorry. 113 00:07:16,080 --> 00:07:16,760 Back we go. 114 00:07:16,920 --> 00:07:24,920 So we create the sort of text message that you need for Autogen agent chat passing in the content of 115 00:07:24,920 --> 00:07:27,000 our message and the source is user. 116 00:07:27,240 --> 00:07:35,080 And this is the onmessage that we call in agent Chat land with this text message and the cancellation 117 00:07:35,080 --> 00:07:36,520 token if you remember that thing. 118 00:07:37,160 --> 00:07:39,840 And then we get back the content. 119 00:07:40,120 --> 00:07:41,880 We say that that was the reply. 120 00:07:41,880 --> 00:07:49,160 And we return a new instance of message of our message, with the content being that reply. 121 00:07:50,120 --> 00:07:53,280 I hope you followed all that, or at least got enough of an idea of it. 122 00:07:53,320 --> 00:07:54,760 Let's give this a shot. 123 00:07:54,920 --> 00:08:00,320 Let's create a new runtime, a single threaded agent runtime, standalone runtime. 124 00:08:00,320 --> 00:08:06,490 We're going to register two agents the simple agent and the LM agent. 125 00:08:06,530 --> 00:08:10,610 The new one, Simple Agent, isn't really an LM at all. 126 00:08:10,610 --> 00:08:12,770 It just disagrees with whatever it's told. 127 00:08:12,770 --> 00:08:21,250 And LM agent is actually going to dispatch through this underscore delegate down to a true gpt4 mini. 128 00:08:21,730 --> 00:08:23,850 So we're going to start the runtime. 129 00:08:24,330 --> 00:08:27,130 Then we are going to send a message. 130 00:08:27,130 --> 00:08:30,410 Hi there to the real agent. 131 00:08:30,610 --> 00:08:33,610 And with whatever it replies we're going to send that. 132 00:08:33,610 --> 00:08:39,890 We're going to forward that on to the other agent, to the simple agent and see what comes out there. 133 00:08:42,210 --> 00:08:44,850 So the LM agent received hi there. 134 00:08:44,850 --> 00:08:46,930 And it said hello, how can I assist you today? 135 00:08:46,970 --> 00:08:49,170 A classic GPT four minion response. 136 00:08:49,570 --> 00:08:53,770 And, uh, the simple agent said, this is simple agent default. 137 00:08:53,770 --> 00:08:55,690 You said, hello, how can I assist you today? 138 00:08:55,690 --> 00:08:59,610 And I disagree, and there you go. 139 00:09:00,050 --> 00:09:04,330 Um, and with that, we will we will stop this pantomime.