WEBVTT 00:00.240 --> 00:00.840 Okay. 00:00.880 --> 00:02.280 It did indeed finish. 00:02.320 --> 00:03.840 It completed its three searches. 00:03.840 --> 00:06.600 It ended with hooray, which is success. 00:06.600 --> 00:08.200 And I received an email. 00:08.200 --> 00:09.160 And here it is. 00:09.160 --> 00:16.560 A nice HTML email, well formatted with a good introduction and overview, detailed analysis crew. 00:16.600 --> 00:19.680 It's now putting at the top, which is in fact what we'll be doing next week. 00:19.680 --> 00:20.760 Look forward to that. 00:21.000 --> 00:24.200 And then there's generally some great information in here. 00:24.200 --> 00:28.040 I would say that there's a couple of misses, but it's generally done pretty well. 00:28.040 --> 00:32.400 And then at the bottom are some references with links since it's an HTML email. 00:32.400 --> 00:36.000 And if I click on that it does appear to work, which is nice. 00:36.240 --> 00:43.240 And I will say, I do find it very satisfying that with this minimal scaffolding with something that 00:43.240 --> 00:45.240 was really very simple indeed. 00:45.280 --> 00:46.840 There's not much code to this look. 00:46.880 --> 00:49.240 That's all there is with just this. 00:49.280 --> 00:53.320 We were able to build this whole deep research framework. 00:53.560 --> 01:00.640 And whilst, you know, maybe those those results are reasonably simplistic, I hope you appreciate 01:00.640 --> 01:05.120 how much potential there is with what we've just built, and can think of all the different ways that 01:05.120 --> 01:06.800 we could expand on this. 01:06.840 --> 01:12.560 And in fact, one very simple way we can expand on this is just to turn up that number, which you may 01:12.560 --> 01:17.600 not want to do, because as I say, it will cost that few cents a piece, but there's no reason why 01:17.600 --> 01:24.200 I can't do it for you so that we can get some to show you what it looks like to have that deeper kind 01:24.200 --> 01:25.080 of research. 01:25.080 --> 01:26.800 So I will do that right now. 01:26.800 --> 01:33.760 We will change that to 20 and, uh, go through and make sure that we, um, hang on if I just set that 01:33.760 --> 01:34.600 properly. 01:35.280 --> 01:38.720 Yes, I have that has now recreated that planner agent. 01:39.000 --> 01:48.440 So now we will come back down here and just make sure that we rerun everything and rerun this. 01:48.480 --> 01:55.240 And now we will expect to do rather a lot more searches, 20 more searches. 01:55.400 --> 02:01.360 And this may take a moment longer and then I will come back and show you what results we get. 02:01.680 --> 02:04.960 And that is completed after 20 searches. 02:05.160 --> 02:10.160 And I can show you the results, which actually it looks generally a bit similar. 02:10.160 --> 02:14.800 It's got the same key frameworks, but it's got more information laid around it. 02:14.840 --> 02:19.320 It has applications and benefits associated with each of the frameworks that are identified. 02:19.320 --> 02:23.440 And then it has some commercial implications at the end and a conclusion. 02:23.440 --> 02:26.160 And so it's certainly more substantive than the previous one. 02:26.160 --> 02:27.080 It's a step up. 02:27.400 --> 02:33.400 And obviously, should you wish, I would very much recommend that you experiment with this. 02:33.520 --> 02:36.600 And we should also, of course, go in and look at our trace. 02:36.800 --> 02:43.360 And if we come in to the research trace, we will see that all of these agents ran and they all ran 02:43.360 --> 02:44.360 in parallel. 02:44.520 --> 02:47.600 The planner agent took took took that time to start with. 02:47.960 --> 02:49.160 Then there's load more. 02:49.880 --> 02:52.280 There's all of this and more. 02:53.320 --> 02:57.250 And then at the end was the writer agent and the email agent. 02:57.410 --> 02:58.930 So you can see everything. 02:58.970 --> 03:05.210 And it definitely gives you that sense of how Asyncio ran all of these different searches in parallel 03:05.210 --> 03:06.290 at the same time. 03:06.410 --> 03:10.290 And then, of course, the writer and email were sequential. 03:10.690 --> 03:16.570 And that then is the conclusion of this part of the deep research agent. 03:17.130 --> 03:22.850 The thing I'd like you to do is bear in mind, have a think about how you could make this more substantive. 03:22.850 --> 03:24.010 What more could be done? 03:24.010 --> 03:27.130 What would you like to do to beef this up? 03:27.410 --> 03:31.290 And then we'll do our quick wrap up before we launch into tomorrow. 03:31.970 --> 03:37.690 So I really hope you enjoyed that as much as I did, and that you appreciate how simple it was to build 03:37.690 --> 03:43.650 that framework around a deep research agent in your deeply puzzling over how you could make it better. 03:43.930 --> 03:48.290 But actually, what we're going to be spending tomorrow doing is putting this into an application and 03:48.290 --> 03:52.090 having it be something that is like a takeaway deep research agent. 03:52.250 --> 03:53.850 And I'm very excited for it. 03:53.850 --> 03:54.850 And I will see you then.