WEBVTT

NOTE Sentence-level transcript of https://www.youtube.com/watch?v=Lrw0jqBNaw0

NOTE One cue per sentence. Cue ids are the line anchors on /transcripts/Lrw0jqBNaw0.html. A cue ends where the next begins, or 2 s after its last word.

s1
00:00:01.309 --> 00:00:03.309
[music]

s2
00:00:12.720 --> 00:00:13.360
Hi everyone.

s3
00:00:13.360 --> 00:00:17.240
Sorry for the start with technical difficulties and all of that.

s4
00:00:17.240 --> 00:00:19.920
Uh, we made it to the end of this track.

s5
00:00:19.920 --> 00:00:21.200
Super exciting.

s6
00:00:21.200 --> 00:00:24.560
Thank you everybody for sticking it out this long.

s7
00:00:24.560 --> 00:00:29.160
Um, are there any developer advocates or devrel people in the audience?

s8
00:00:29.160 --> 00:00:30.440
Raise your hand.

s9
00:00:30.440 --> 00:00:31.440
Yeah, okay.

s10
00:00:31.440 --> 00:00:34.960
So did you come to like throw tomatoes at me cuz I'm talking about the dead now.

s11
00:00:34.960 --> 00:00:38.960
Okay, so it's not going to be all doom and gloom like that.

s12
00:00:38.960 --> 00:00:41.400
Um, a bit of like backstory in this.

s13
00:00:41.400 --> 00:00:43.000
Um, I'm a research scientist.

s14
00:00:43.000 --> 00:00:46.080
So last year I was an astronomer.

s15
00:00:46.080 --> 00:00:47.320
Um, and I just sort of like wound up.

s16
00:00:47.320 --> 00:00:49.000
I didn't know what GTM was or any of that.

s17
00:00:49.000 --> 00:00:51.000
I just sort of wound up in this.

s18
00:00:51.000 --> 00:01:00.320
Um, and I submitted like a bunch of boring sciency eval talks that were unceremoniously I I assumed thrown into the trash uh, for this conference.

s19
00:01:00.320 --> 00:01:09.880
But my manager, who is a developer advocate, he put in, you know, the death the death of developer advocates, which is, you know, appropriately buzzworthy and hypey.

s20
00:01:09.880 --> 00:01:11.320
And so so that was great.

s21
00:01:11.320 --> 00:01:18.720
But his title is developer advocate, so it didn't really necessarily make as much sense for him to be coming up here and giving his eulogy.

s22
00:01:18.720 --> 00:01:25.960
So we brainstormed like maybe I would dress up as like a robot and like a maul him and attack him on the stage or something like that.

s23
00:01:25.960 --> 00:01:29.960
Um, but then it just like logistically it was going to be hard to do that.

s24
00:01:29.960 --> 00:01:31.880
Uh, so he just went on vacation.

s25
00:01:31.880 --> 00:01:46.800
Uh, so I'm here uh, as the agent advocate uh, to talk about this sort of like new role and uh, try to advocate for it and uh, convince all of you that we should all be agent advocates to help

s26
00:01:46.800 --> 00:01:48.560
uh, in this new era.

s27
00:01:48.560 --> 00:01:59.240
So uh, zooming out a little bit and going back uh, in time a bit because uh, I was trying to talk about developer advocates to somebody at the conference yesterday and their eyes like glazed over they had no idea what I was talking about.

s28
00:01:59.240 --> 00:02:03.880
So just to sort of talk about what what this thing is that I'm saying is dead.

s29
00:02:03.880 --> 00:02:05.360
Uh so back in the '80s, right?

s30
00:02:05.360 --> 00:02:13.000
It was called like software evangelism where one would go forth and speak the good word of the product and bring it out there.

s31
00:02:13.000 --> 00:02:29.160
But then fast forward to the 2010s or so, that's when developer advocacy advocacy started to become a thing where now instead of having this single trajectory of the communication pathway, now it's a feedback loop and a two-way street where you have these people with very deep empathy for developers

s32
00:02:29.160 --> 00:02:37.280
who understand them and speak their language and could understand um what their needs were um and then bring that back to the product.

s33
00:02:37.280 --> 00:02:39.920
And then um these developers, right?

s34
00:02:39.920 --> 00:02:46.040
Fast forward even more, they have so much influence within their company and basically become these like kingsmakers.

s35
00:02:46.040 --> 00:02:52.320
Uh and so the developer experience became a very important aspect of the go-to-market sort of strategy.

s36
00:02:52.320 --> 00:03:00.959
Um but now in 2026, uh developers are no longer working alone and what it means to be a developer is completely changing.

s37
00:03:00.959 --> 00:03:07.840
Um and so our role, right, as developer advocates um developer in developer relations, we're relating to developers.

s38
00:03:07.840 --> 00:03:15.560
And so as the role of developers fundamentally changing, so must then does the role of the developer advocate.

s39
00:03:15.560 --> 00:03:20.800
Um so in this slide I'm just kind of talking about the other users, right?

s40
00:03:20.800 --> 00:03:24.600
So what's happening uh with DevRel uh outside of the agent.

s41
00:03:24.600 --> 00:03:27.320
So most of the talk is going to be talking about the agent as a user.

s42
00:03:27.320 --> 00:03:38.320
But I also did did want to bring up, right, that engineers they're becoming like these orchestrators of these fleets of agents, um babysitters and whatnot of these things.

s43
00:03:38.320 --> 00:03:42.920
Um and their job, like all of the job postings and whatnot, there's language is continuously changing, right?

s44
00:03:42.920 --> 00:03:46.120
They're um expected to have this AI fluency.

s45
00:03:46.120 --> 00:03:53.000
Um and at the same time, there's also, you know, people like me, like uh non-engineers, right?

s46
00:03:53.000 --> 00:03:54.320
I was a research scientist.

s47
00:03:54.320 --> 00:04:01.360
I had like zero commits on GitHub last year, and now I have 12,000, and I'm like an open source maintainer for multi-agent orchestration framework.

s48
00:04:01.360 --> 00:04:09.160
Like, we have so much like capability now with all of these agents, and now anybody with these agents can use dev tools, essentially.

s49
00:04:09.160 --> 00:04:18.320
So, you have this whole other persona and ICP uh to potentially be relating to and um having empathy with when you're there using your product.

s50
00:04:19.359 --> 00:04:24.840
So, let's talk about now this whole new user that we have in the form of an agent.

s51
00:04:24.840 --> 00:04:28.680
Um so, an agent is somewhat unique, right?

s52
00:04:28.680 --> 00:04:34.160
In the sense that it is both the user of your tool in a very similar way to the developer.

s53
00:04:34.160 --> 00:04:37.919
It's going out reading your docs, but it's just reading them differently cuz it's a machine.

s54
00:04:37.919 --> 00:04:39.240
Um you know, it's calling the API.

s55
00:04:39.240 --> 00:04:44.680
It's encount- it's ha- has its own frustrations with how it's encountering errors and recovering from them, right?

s56
00:04:44.680 --> 00:04:47.440
But then it's also a recommender of your tools.

s57
00:04:47.440 --> 00:04:48.760
Um but somewhat similar, right?

s58
00:04:48.760 --> 00:04:53.680
To developers in the way that they are also recommenders of your tools in a more organic, bottom-up way.

s59
00:04:53.680 --> 00:04:56.640
Um so, the whole, you know, basis for DevRel, right?

s60
00:04:56.640 --> 00:04:58.960
Is to encourage that bottom-up adoption.

s61
00:04:58.960 --> 00:05:04.480
But now the adoption and the recommendation system, a lot of it's being driven by the agent itself.

s62
00:05:04.480 --> 00:05:18.480
That is either, you know, maybe servicing your product directly through like ChatGPT or Claude, like directly in a Q&A sort of environment, or it's, as we had heard like in some of the previous talks where the speaker asked folks like, "How many of you have just let your agent

s63
00:05:18.480 --> 00:05:19.880
install a library for you?"

s64
00:05:19.880 --> 00:05:21.680
And like, there were many hands went up, right?

s65
00:05:21.680 --> 00:05:32.360
So, there's this like recommender of tools where basically it's just installing these like frameworks and things um directly and embedding them into the workflow um and sort of working with the developer

s66
00:05:32.360 --> 00:05:34.800
um in that taste.

s67
00:05:36.080 --> 00:05:37.520
So, I know it's late for numbers.

s68
00:05:37.520 --> 00:05:40.600
You don't have to read them or anything like that.

s69
00:05:40.600 --> 00:05:44.919
Um so, I have a couple different concrete examples for measuring these seats, right?

s70
00:05:44.919 --> 00:05:48.560
Cuz I am a data science scientist nerd person.

s71
00:05:48.560 --> 00:05:58.360
Um so one of my first projects when I was uh working on this um uh when I became an agent advocate was to build um a benchmark called CodeScaleBench.

s72
00:05:58.360 --> 00:06:02.880
And so I developed hundreds of tasks that were reflective of the software development life cycle.

s73
00:06:02.880 --> 00:06:07.160
And I basically unleashed these agents with and without um our product tooling.

s74
00:06:07.160 --> 00:06:10.720
So I work at Sourcegraph and we have a code navigation MCP tool.

s75
00:06:10.720 --> 00:06:18.760
Um and the point of that was to understand, okay, how is our tool helping the agent do the work that it's, you know, going to be doing.

s76
00:06:18.760 --> 00:06:21.680
Um and when it isn't working well, why isn't it working well?

s77
00:06:21.680 --> 00:06:24.160
So that we can then go in and actually fix that.

s78
00:06:24.160 --> 00:06:26.640
Um so I have thousands and thousands of these traces.

s79
00:06:26.640 --> 00:06:37.000
And I I as we have heard in like the previous talks, like now we have these amazing logs of data for like these really tight feedback loops where you can see exactly where it's breaking down and then go in and fix it.

s80
00:06:37.000 --> 00:06:43.480
Uh so this one specific example here was um when I was looking at how it was like using a read tool.

s81
00:06:43.480 --> 00:06:52.919
Um and the model had the these expectations based off of its like biases from how it from its training data of what it expected for a particular um command

s82
00:06:52.919 --> 00:06:55.400
um that would be available within the tool.

s83
00:06:55.400 --> 00:06:59.480
And there's nothing in our description uh that would have like led it to believe otherwise.

s84
00:06:59.480 --> 00:07:03.919
So it tried to use like read line instead of start line or something like that.

s85
00:07:03.919 --> 00:07:07.919
And then it ended up failing, but then at least the error told it why it failed.

s86
00:07:07.919 --> 00:07:09.800
So it was like, okay, that that was a good part of it.

s87
00:07:09.800 --> 00:07:11.720
So it was able to fix itself.

s88
00:07:11.720 --> 00:07:15.000
But then it's burning right an entire turn just failing.

s89
00:07:15.000 --> 00:07:19.560
And you could just go in and fix that um aspect of like how it's interacting with the tool.

s90
00:07:19.560 --> 00:07:30.360
And this is really important, right, to gather that feedback um and understand the friction that like now your new agent user is having with your tool because it's the way that um different organizations are going to be evaluating your tool, right?

s91
00:07:30.360 --> 00:07:36.080
In terms of not just is it working well, but like how many tokens is the agent dealing with to work with your tool?

s92
00:07:36.080 --> 00:07:37.720
And how fast is it?

s93
00:07:37.720 --> 00:07:44.800
Um so this is, you know, really an important aspect of the role is measure um, how these users are using it.

s94
00:07:44.800 --> 00:07:48.120
The other side of it um, is like the recommendation layer, right?

s95
00:07:48.120 --> 00:07:50.880
So, the uh, GEO instead of SEO.

s96
00:07:50.880 --> 00:07:54.120
So, the generative engine optimization.

s97
00:07:54.120 --> 00:07:59.160
Um, and I didn't mention it before, but in the previous slide um, I had a GitHub repo.

s98
00:07:59.160 --> 00:08:01.640
Like, there's two different toy projects that I put together.

s99
00:08:01.640 --> 00:08:07.000
At the end of the talk, there's like a QR code with a link that you can send your agent to to like have access to all this.

s100
00:08:07.000 --> 00:08:13.200
So, don't worry about like taking screenshots All of all of the data will be released to you.

s101
00:08:13.200 --> 00:08:15.480
Um, so anyway, back to this.

s102
00:08:15.480 --> 00:08:18.520
Um, I set up a little experiment, right?

s103
00:08:18.520 --> 00:08:26.320
To see how uh, these different chatbots and agents and whatnot were recommending our product or like mentioning it at all.

s104
00:08:26.320 --> 00:08:36.760
Um, and so there's a, you know, process to that cuz you have you want to understand like, what is your ICP actually doing when you would want your product to be surfaced?

s105
00:08:36.760 --> 00:08:39.400
So, there was a bit of a gap that I found.

s106
00:08:39.400 --> 00:08:53.000
Um, if I had designed some of these prompts around somebody who like was actively shopping for this sort of code intelligence sort of tooling and doing a comparative sort of thing, then our product was ending up being recommended like 65%

s107
00:08:53.000 --> 00:08:54.040
of the time.

s108
00:08:54.040 --> 00:09:06.920
Um, but what I found was the arguably like the more typical use case and where we'd want to be showing up for people when they're encountering a specific pain or have a specific need where our product could serve them better,

s109
00:09:06.920 --> 00:09:09.000
uh, zero mentions, right?

s110
00:09:09.000 --> 00:09:18.560
So, in this particular instance, um, I put in a prompt that was like, we keep breaking downstream services when we change shared libraries because we can't see all the consumers.

s111
00:09:18.560 --> 00:09:25.080
And you know, our one uh, part of our product is being able to have this observability layer to like see across all the repos.

s112
00:09:25.080 --> 00:09:32.960
So, we'd want uh, some level of like attribution or recognition from um, an agent to say, "Hey, you could use something like this."

s113
00:09:32.960 --> 00:09:38.200
But instead it said, uh, "You could just have your developers make a wiki page or something.

s114
00:09:38.200 --> 00:09:45.400
Um but with this, you know, we wouldn't know that without running these sorts of experiments um and getting this sort of data.

s115
00:09:45.400 --> 00:09:55.040
So, what this leads to is like then you can have a hypothesis of okay, maybe the messaging that we're putting out there isn't uh attributing some of these pains and use cases clearly enough

s116
00:09:55.040 --> 00:09:56.840
for the agents to be picking it up.

s117
00:09:56.840 --> 00:10:15.080
So, we have uh like a content campaign in the works to um make changes to our website and then we can directly measure whether that has like an actual lift and not necessarily in the form of like anything that was baked into the training data, but then how uh the agents that are using those like web search

s118
00:10:15.080 --> 00:10:21.120
tool calls, how they are then interpreting um the information about your product.

s119
00:10:21.400 --> 00:10:35.560
So, you know, there are just some um different ways that you could think about guiding the agents um to help support like the servicing, the discoverability of your product and this user finding it um at their moment of need, right?

s120
00:10:35.560 --> 00:10:41.520
Um so, for example, um this whole field is moving so fast.

s121
00:10:41.520 --> 00:10:44.720
Uh so, I mean, training data is always going to be stale.

s122
00:10:44.720 --> 00:10:56.360
Actually, in the um GEO pilot study that I did, the data that I was showing there, that was using Claude Sonnet 4. It's very old um obviously and I just today, this afternoon, ran it with 4.6

s123
00:10:56.360 --> 00:10:59.240
thinking that okay, surely it's going to it's going to be better.

s124
00:10:59.240 --> 00:11:08.680
It's going to know like improved information about our product, but uh so, in the previous model, it kept pitching Cody, which was like one of our older products.

s125
00:11:08.680 --> 00:11:14.120
Um but if I when I uh ran it again, it it pitched Cody even more, right?

s126
00:11:14.120 --> 00:11:21.360
Cuz like now you have all of these like old models like uh outputting content that then is like compounding in the internet.

s127
00:11:21.360 --> 00:11:27.000
So, you have to figure out like how to bury all of that uh noise with your true signal.

s128
00:11:27.000 --> 00:11:35.720
Um and the way that some folks are working on that is as we've heard from other people like these LLMs at TXT uh sort of pages, right?

s129
00:11:35.720 --> 00:11:40.920
So, you have more authoritative sources of truth that you're hoping to direct the agent to.

s130
00:11:40.920 --> 00:11:48.440
But, they still need to be using the tools and using real-time information and provenance to be able to give accurate answers about your product.

s131
00:11:48.440 --> 00:11:51.160
You also want to give like the agent something to quote, right?

s132
00:11:51.160 --> 00:11:56.280
They they they want to bring something that they can really sell to the to the user, right?

s133
00:11:56.280 --> 00:11:59.160
So, you want current examples and keep everything up-to-date.

s134
00:11:59.160 --> 00:12:03.160
Like, even if your stuff hasn't changed in 2 years, which would be shocking.

s135
00:12:03.160 --> 00:12:10.240
Even if it hasn't, like keep everything up-to-date and fresh because that, you know, part of that is how they have their relevance algorithm.

s136
00:12:10.240 --> 00:12:14.120
And they also really really like charts and FAQs and things like that.

s137
00:12:14.120 --> 00:12:18.320
And you also want to make sure your product is where the agents are, right?

s138
00:12:18.320 --> 00:12:19.960
You're going to market.

s139
00:12:19.960 --> 00:12:22.080
So, go go to agent market, right?

s140
00:12:22.080 --> 00:12:28.680
So, make sure you're in the marketplace in the MCP registries, everywhere that you would expect an agent to be able to easily find you.

s141
00:12:28.680 --> 00:12:40.200
And also make sure that you know, that whole you reduce as much friction as possible for an agent or and developer to go from finding out about your tool to embedding it in their workflow.

s142
00:12:40.200 --> 00:12:52.520
Because if an agent realizes your tool requires like three different demos and emailing sales reps and stuff, they're never going to say, "Hey user, like here's what you should do, but FYI, you're going to have to do all this other stuff."

s143
00:12:52.520 --> 00:12:54.120
It's like not going to happen.

s144
00:12:54.120 --> 00:12:58.560
And then also make sure that you are covering that those pains, right?

s145
00:12:58.560 --> 00:13:03.560
Because that's how a user is going to be most like in their time of need, right?

s146
00:13:03.560 --> 00:13:08.680
That's going to be the best opportunity for your product and your service, right, to be surfaced to them.

s147
00:13:08.680 --> 00:13:17.520
And so, you want to make sure that there's enough content out there on the internet for the agent to like be aware of that and make those connections for you.

s148
00:13:17.839 --> 00:13:28.480
And so, right, there's this like ongoing question of what even the heck is DevRel and advocacy and now now this agent advocacy thing, right?

s149
00:13:28.480 --> 00:13:29.360
So like where does it fit?

s150
00:13:29.360 --> 00:13:30.240
Where does it go?

s151
00:13:30.240 --> 00:13:31.320
Like is it engineering?

s152
00:13:31.320 --> 00:13:32.000
Is it product?

s153
00:13:32.000 --> 00:13:32.760
Is it marketing?

s154
00:13:32.760 --> 00:13:34.600
It's like yeah, yes, yes.

s155
00:13:34.600 --> 00:13:36.240
It's all of those things.

s156
00:13:36.240 --> 00:13:41.080
And and with the rise of agents it hasn't gotten any clearer, right?

s157
00:13:41.080 --> 00:13:43.040
Those seams haven't gotten any clearer.

s158
00:13:43.040 --> 00:13:47.840
If anything though, everybody's role with across the organization has gotten fuzzier.

s159
00:13:47.840 --> 00:13:50.280
So that actually helps in a lot of ways.

s160
00:13:50.280 --> 00:13:56.400
Um and but you can sort of split it up and think about it in terms of like these different flavors, right?

s161
00:13:56.400 --> 00:14:04.240
And you can mix and match depending on whatever skills and abilities various employees have within your organization and whatever the product needs at a given time.

s162
00:14:04.240 --> 00:14:06.600
So you have like the engineering flavor, right?

s163
00:14:06.600 --> 00:14:18.120
And those are folks that are partnering directly with the engineering team to make these interfaces for how the agent is talking to your product like through the MCP server and building out these evals and the instrumentation.

s164
00:14:18.120 --> 00:14:19.160
Then you have the product flavor.

s165
00:14:19.160 --> 00:14:22.240
So those are folks that are going to own the end-to-end agentic experience, right?

s166
00:14:22.240 --> 00:14:31.600
And so translating these evals to bring it to the product team and like having the agent experience rubrics how they're encountering all of that content.

s167
00:14:31.600 --> 00:14:32.920
And then you have the marketing flavor, right?

s168
00:14:32.920 --> 00:14:46.680
And that should be the folks that are really owning that pipe gen and how the agents are like entering the funnel and finding out about your product and then bringing the developers along with them by surfacing those recommendations.

s169
00:14:48.120 --> 00:14:53.320
So I know I you know said the death of developer advocates.

s170
00:14:53.320 --> 00:14:56.720
But the core right of DevRel still holds.

s171
00:14:56.720 --> 00:15:00.280
It's just you have a change in your audience.

s172
00:15:00.280 --> 00:15:04.360
So it's still extremely important to do enablement, right?

s173
00:15:04.360 --> 00:15:07.280
It's just the type of enablement is a bit different.

s174
00:15:07.280 --> 00:15:14.440
You're educating developers now who are have a completely different type of job where they're orchestrating these fleets of agents.

s175
00:15:14.440 --> 00:15:17.120
And you're also educating agents, right?

s176
00:15:17.120 --> 00:15:26.320
So you're having to put out content that is machine readable, has like agent friendly APIs, all of these things to make it as easy as possible to use your product both for human developers

s177
00:15:26.320 --> 00:15:28.440
and for the agents that they're using.

s178
00:15:28.440 --> 00:15:31.720
And community is also more important than ever, right?

s179
00:15:31.720 --> 00:15:41.800
Um having that human-to-human connection um where developers can come um and uh bring their agents also into the loop, right?

s180
00:15:41.800 --> 00:15:51.880
So that's another component um that needs to be considered uh when you're building these different communities because there's all these questions, right, of privacy and like data concern as well.

s181
00:15:51.880 --> 00:15:57.840
If people are like bringing their Claude's and whatnot like into the Discord and they're like uh recording all of the conversations and everything like this.

s182
00:15:57.840 --> 00:16:00.880
It's just like a new thing they have to think of as a community builder.

s183
00:16:00.880 --> 00:16:09.720
And then there's the feedback loop, so you're still uh responsible for bringing the voice of the developer who's using the agents back to the organization, but then you can also

s184
00:16:09.720 --> 00:16:19.520
uh basically spin up like thousands of these agents to perform experiments on them and experiments that you can't really like do as easily with the developers who don't want to maybe talk to you that much.

s185
00:16:19.520 --> 00:16:20.880
Um and then credibility, right?

s186
00:16:20.880 --> 00:16:26.240
So you need to be earning credibility both from human developers.

s187
00:16:26.240 --> 00:16:30.800
Um so like don't like not using Claude's slop at them, right?

s188
00:16:30.800 --> 00:16:33.560
Then tell your AEs to stop that as well.

s189
00:16:33.560 --> 00:16:36.920
Nobody Everybody knows what it is and nobody likes it.

s190
00:16:36.920 --> 00:16:42.280
Um and but then credibility like actually Claude loves its own slop uh for whatever reason.

s191
00:16:42.280 --> 00:16:45.120
So there's a bias, right, from agents of their own content.

s192
00:16:45.120 --> 00:16:52.040
So whenever you're making like agent-facing content, as long as it's structured, you can have as many m dashes and whatever as as it wants.

s193
00:16:52.040 --> 00:16:57.720
Um but it's just a completely different sort of uh credibility landscape, humans versus agents.

s194
00:16:57.720 --> 00:17:02.200
So what I'm advocating for here, right, is like building out a curb cut.

s195
00:17:02.200 --> 00:17:07.520
So curb cuts were built for wheelchairs, like built for a specific user to use them.

s196
00:17:07.520 --> 00:17:10.600
Um but now everybody, you know, benefits from that, right?

s197
00:17:10.600 --> 00:17:14.800
Anybody with wheels, right, strollers and um suitcases and all of those things.

s198
00:17:14.800 --> 00:17:20.680
So my argument is that by serving the uh agents, uh the human path gets cleared, too.

s199
00:17:20.680 --> 00:17:28.400
There's just, you know, there's just one more user in the room now, but they are still serving the human on the other end, and we're all working together on this.

s200
00:17:28.400 --> 00:17:37.560
So, for, you know, DevRel, one quick thing that you could do like right away is point a coding agent at your docs, and then looking through that transcript and start developing your agent experience report.

s201
00:17:37.560 --> 00:17:48.960
And then if you're more on the GTM side, start like developing some of these experiments with the GEO, putting together those prompts, and looking at the mentions versus recommendations.

s202
00:17:48.960 --> 00:17:52.320
And I made this whole talk agent legible, right?

s203
00:17:52.320 --> 00:17:58.440
So, there's a QR code there, as well as a couple different toy repos that have some templates for you to get started.

s204
00:17:58.440 --> 00:18:00.640
And that's it.

s205
00:18:13.862 --> 00:18:15.862
[music]