WEBVTT 00:00.240 --> 00:03.280 Well, hello and welcome to week three. 00:03.320 --> 00:04.240 We are here. 00:04.240 --> 00:13.160 We are at Crew Week, and this is the time that I get to unveil the world of crew A, and I want to 00:13.160 --> 00:18.680 say that you might find it a little bit painful to make this mental change now, because we've just 00:18.680 --> 00:21.080 fallen in love with OpenAI agents SDK. 00:21.120 --> 00:23.400 We've got used to it, we know all about it. 00:23.400 --> 00:27.440 And suddenly there's going to be a change new terminology, new constructs. 00:27.440 --> 00:30.440 And the thing is, I would say there's a lot in common. 00:30.440 --> 00:34.320 There's going to be some differences too, but you're going to find you're quickly going to fall in 00:34.320 --> 00:35.880 love with crew AI as well. 00:35.920 --> 00:40.040 And this is going to be a repeating pattern in the next few weeks that we're going to get comfortable 00:40.040 --> 00:44.040 with something and love it, and then we're going to have to put it to one side and move on to the next 00:44.040 --> 00:45.280 and just keep it in mind. 00:45.320 --> 00:47.160 Keep keep in mind the differences. 00:47.160 --> 00:50.480 What's similar, what's unique, what's better, what's worse. 00:50.480 --> 00:54.320 Because different projects will lend themselves more to different frameworks. 00:54.320 --> 00:56.160 And you're going to find that you'll fall in love with one. 00:56.160 --> 00:59.440 And it may not be the same as me, and you should be able to make their own determination. 00:59.440 --> 01:01.950 And you'll learn something from each of these experiments. 01:01.990 --> 01:04.910 Plus all of our commercial projects will be a bit different. 01:05.110 --> 01:06.630 Okay, that's enough preamble. 01:06.630 --> 01:07.430 Let's get to it. 01:08.190 --> 01:11.510 So crew is actually several different things. 01:11.510 --> 01:13.470 We're gonna we're gonna have this a couple of times. 01:13.510 --> 01:14.950 Autogen is much the same. 01:14.990 --> 01:19.550 But crew in particular, when you when you hear people talk about crew, they might be talking about 01:19.590 --> 01:27.350 something called crew I enterprise, which is crews platform for deploying agents, for running them, 01:27.350 --> 01:30.910 for monitoring and managing them through a number of different screens. 01:31.350 --> 01:33.670 They sometimes just call it the career AI platform. 01:33.710 --> 01:37.430 Sometimes I think officially it's branded crew AI enterprise. 01:37.470 --> 01:45.230 And if you go to their landing page at currycomb, not crew Currycomb, you'll see that this is what 01:45.270 --> 01:47.670 sort of presented first and foremost. 01:48.390 --> 01:56.670 Secondly, there is a product called crew crew I, UI studio, which is one of these low code, no code 01:56.670 --> 02:00.070 platforms for piecing together agent interactions. 02:00.190 --> 02:02.950 A bit like an addendum that we looked at at the very, very beginning. 02:03.390 --> 02:06.910 But it's a it's a nice, elegant end user tool. 02:07.150 --> 02:13.910 And then thirdly, it is something called the framework, the crew AI framework, which is an open source 02:13.910 --> 02:15.190 framework for. 02:15.230 --> 02:21.310 And I quote from, from the site orchestrating high performing AI agents with ease and scale. 02:21.710 --> 02:25.150 So these are the three different offerings that they have. 02:25.470 --> 02:30.710 And hopefully no surprise to you, we are going to be focusing on the open source framework because 02:30.710 --> 02:32.750 we're here to build agents ourselves. 02:32.750 --> 02:34.110 We're writing the code. 02:34.110 --> 02:39.470 We're not going to be using the low code tooling, which is there, and we're also not going to be necessarily 02:39.470 --> 02:45.350 needing to do something where we're deploying them and paying for a hosting platform, which is really 02:45.350 --> 02:47.150 crew AI enterprise. 02:47.550 --> 02:53.270 But it's perhaps worth pointing out that when we think of the differences between these different platforms, 02:53.550 --> 02:58.790 OpenAI and anthropic have the great benefit that they already have a reason for being. 02:58.830 --> 03:04.620 They they have their models and that is their source of revenue when it comes to groups like cry. 03:04.660 --> 03:08.580 Of course, they they need to be very mindful of a monetization strategy. 03:08.740 --> 03:12.180 And the open source framework is very popular, very successful. 03:12.180 --> 03:16.340 But of course, they also need ways to monetize and cry enterprise. 03:16.340 --> 03:22.820 And the their tooling is their path to doing that, which makes complete sense and which I don't I don't 03:22.820 --> 03:24.220 hold it against them for a moment. 03:24.220 --> 03:28.940 But it does, of course, mean that when you go to things like their website, there is a lot of upselling 03:28.940 --> 03:35.020 going on that they want to try and win people over so that not only will they use the open source toolkit, 03:35.020 --> 03:42.060 but then they'll end up paying for hosting and deployment within the broader cry platform. 03:42.980 --> 03:47.860 And so now for the rest of what we talk about, it will always be the open source framework that we 03:47.860 --> 03:49.300 will be working with. 03:49.340 --> 03:54.660 And when you work with the framework, there are in fact two different flavors, two different approaches 03:54.660 --> 03:57.860 that you can use for all of your work with the framework. 03:57.900 --> 04:05.500 And one of them is called crew I Crews, which is when you have autonomous solutions with teams working 04:05.540 --> 04:07.660 together, agents of different roles. 04:08.020 --> 04:12.380 Crew is crews word for a sort of team of agents, a crew of agents. 04:12.940 --> 04:16.060 And then there's also something called crew I flows. 04:16.380 --> 04:21.700 And this I think this is actually a newer part of crew because at least I wasn't aware of it about six 04:21.700 --> 04:23.180 months ago when I last used crew. 04:23.220 --> 04:24.140 It may have been there, but. 04:24.140 --> 04:28.100 And I just didn't notice it, but but it's certainly now quite prominently in the documentation. 04:28.340 --> 04:36.660 Cry flows are more, um, prescribed workflows, fixed workflows where you divide a problem into multiple 04:36.660 --> 04:42.340 steps and you have workflows that lead through it with decision points and outcomes and so on. 04:42.860 --> 04:48.060 And their their documentation suggests that you should choose crews when you're looking for autonomous 04:48.060 --> 04:56.180 problem solving, creative collaboration, or exploratory tasks versus flows, which is more about deterministic 04:56.220 --> 05:00.180 outcomes, auditability or precise control. 05:00.300 --> 05:06.690 And I probably I imagine and this is just speculation, but I'm imagining that flows has come out as 05:06.690 --> 05:13.650 a result of some of the concerns people have about running crews in production, where there is this 05:13.650 --> 05:21.690 greater level of uncertainty and lack of auditability and sometimes a tighter defined flow, a work 05:21.730 --> 05:26.770 flow is what's required rather than a fully agentic autonomous solution. 05:27.570 --> 05:31.410 So that hopefully gives you a bit of context, and again, no surprise to know that we're going to be 05:31.410 --> 05:34.650 focusing on the crews, because this is all about agents. 05:34.690 --> 05:40.890 This course and building workflows is something that you can also do, but it's a bit more straightforward. 05:40.890 --> 05:46.690 And of course, you could also do that simply by calling LMS directly and by interpreting their responses. 05:46.810 --> 05:51.010 What we're interested in is the more autonomous aspect of this. 05:51.050 --> 05:56.850 It is about when we allow different LMS to choose their own adventure and to go about solving their 05:56.850 --> 05:58.850 problems in an agentic way. 05:59.210 --> 06:01.890 So that will be our focus for this week.