WEBVTT

NOTE Sentence-level transcript of https://www.youtube.com/watch?v=7wu2hsRfvV0

NOTE One cue per sentence. Cue ids are the line anchors on /transcripts/7wu2hsRfvV0.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
00:00:13.100 --> 00:00:15.100
[applause]

s3
00:00:15.560 --> 00:00:17.520
Coming guys, can you hear me just fine?

s4
00:00:17.520 --> 00:00:18.720
All good?

s5
00:00:18.720 --> 00:00:19.320
Okay, awesome.

s6
00:00:19.320 --> 00:00:28.600
Just so I can contextualize this talk a little bit, can I get a show of hands of who here is an engineer or is a forward and in a forward deployed motion at all?

s7
00:00:28.840 --> 00:00:29.480
Okay.

s8
00:00:29.480 --> 00:00:30.680
Okay, so I'm in my minority.

s9
00:00:30.680 --> 00:00:31.560
Okay, awesome.

s10
00:00:31.560 --> 00:00:32.640
Uh sounds good.

s11
00:00:32.640 --> 00:00:34.320
So yes, um I'm Sunny.

s12
00:00:34.320 --> 00:00:37.600
I'm the uh CTO of Forward Deployed Engineering here at Decagon.

s13
00:00:37.600 --> 00:00:48.080
And today I'll talk about what it is that we do, why we have a forward deployed motion, how it has changed over time as we've gone from 50 people to 500 people over the course of a year.

s14
00:00:48.080 --> 00:00:54.120
Um how it changes if you're working with a Fortune 20 versus a more mid-market brand.

s15
00:00:54.120 --> 00:00:55.520
Uh thank thank you all for coming.

s16
00:00:55.520 --> 00:00:56.520
I hope it's useful.

s17
00:00:56.520 --> 00:00:58.840
And um yeah, let's get started.

s18
00:00:58.840 --> 00:01:02.560
So, Okay.

s19
00:01:02.560 --> 00:01:06.200
Uh okay, so just to give context on what Decagon is.

s20
00:01:06.200 --> 00:01:11.920
Um for those of you unfamiliar, Decagon is a 24/7 AI customer service agent.

s21
00:01:11.920 --> 00:01:27.040
So, we've all had the experience of calling into your favorite brand and being told to press one for billing, press two for membership options, etc. Or you email into your brand because you need urgent support and you hear back in two or three business days.

s22
00:01:27.040 --> 00:01:28.400
Decagon replaces all of that.

s23
00:01:28.400 --> 00:01:34.720
So, instead you pick up and you call your brand of choice and you get a human-like agent who is helping you.

s24
00:01:34.720 --> 00:01:37.880
You email in, you get a human-like reply right away.

s25
00:01:37.880 --> 00:01:41.040
So, that's what Decagon does in a nutshell.

s26
00:01:41.040 --> 00:01:56.160
Um multilingual, omni-channel, etc. Um and then importantly, and again I you know I I I I say this to to help contextualize what our forward deployed motion does.

s27
00:01:56.160 --> 00:02:04.880
But, you know, we land in our customers to help them with the kinds of complex support workflows that today have to go to humans.

s28
00:02:04.880 --> 00:02:15.040
Um but once we are there and our agent is learning about the customers and has a relationship with the customers, then we also work with our customers to figure out, "Hey, how can we actually make you more money?"

s29
00:02:15.040 --> 00:02:19.480
So, one example, you'll see this in the bottom of the slide here, but Hertz, you know, we're all familiar with.

s30
00:02:19.480 --> 00:02:26.680
Hertz came to us because they had these kind of complex inbound support workflows that indeed it to be offloaded it to an agent.

s31
00:02:26.680 --> 00:02:32.160
But once we were there, uh it turns out, "Hey, we already have like these integrations to your back-end systems.

s32
00:02:32.160 --> 00:02:34.120
What other communications are you doing with customers?"

s33
00:02:34.120 --> 00:02:44.440
And one that Decagon now does for them is to reach out proactively to a customer when it's time to renew their car lease or extend it or whatever, and they can do that from within Decagon.

s34
00:02:44.440 --> 00:02:54.520
So, it is you land We typically we land and help them deflect these sort of inbound support cases, and then we expand into, you know, how do we make you more money?

s35
00:02:56.120 --> 00:02:59.880
Um Now, we work cross-vertical.

s36
00:02:59.880 --> 00:03:04.040
We also have really large enterprises, more mid-market brands.

s37
00:03:04.040 --> 00:03:07.440
These are a subset of what I was approved to talk about.

s38
00:03:07.440 --> 00:03:11.200
Uh there were way more I wanted to add in there, but our our marketing had got mad at me.

s39
00:03:11.200 --> 00:03:18.040
We have, you know, the the top left we have our financial institutions, the bottom right we have our, you know, your favorite tech brand.

s40
00:03:18.040 --> 00:03:32.640
Uh and this is relevant because, as I'll talk about briefly, uh the kind of forward deployment you have to do is vastly different based on both the size of the enterprise and also the vertical.

s41
00:03:33.560 --> 00:03:34.040
Okay.

s42
00:03:34.040 --> 00:03:43.880
So, I imagine this is the case for a lot of agentic companies, but Decagon has effectively two kinds of forward deployment engineering.

s43
00:03:43.880 --> 00:03:50.360
Number one is taking that AI customer service agentic brain and making it work for your enterprise.

s44
00:03:50.360 --> 00:04:01.840
So, the same way that you train a human, you give it instructions on what to do when a user asks X, how you respond back to it, what sort of brand tonality you have,

s45
00:04:01.840 --> 00:04:04.800
what actions do you take on behalf of the user.

s46
00:04:04.800 --> 00:04:16.000
All of this like configuring of that human of the of that [clears throat] agent brain is one form of our forward deployment motion, where we work with the customer, we figure out what does success look like for you, how do you want the agent to speak,

s47
00:04:16.000 --> 00:04:20.239
what sort of user intents do you actually want the agent to hand off to a human instead.

s48
00:04:20.239 --> 00:04:22.640
That's the left half of this diagram.

s49
00:04:22.640 --> 00:04:27.720
And we have a team, which I'll talk about briefly, who is like really good at configuring the agent.

s50
00:04:27.720 --> 00:04:30.800
Largely, this can also happen within the UI.

s51
00:04:30.800 --> 00:04:37.640
And then on the right side is um the the previous speaker alluded to this as well.

s52
00:04:37.640 --> 00:04:43.720
Forward deployment forward deployment engineers are the front line for customer product asks.

s53
00:04:43.720 --> 00:04:48.720
And it is their job to figure out, "Hey, enterprise A made this ask.

s54
00:04:48.720 --> 00:04:52.400
I know in 2 weeks enterprise B is also going to have the same ask."

s55
00:04:52.400 --> 00:04:55.040
And this happens with stunning regularity.

s56
00:04:55.040 --> 00:05:02.680
So, I want to make sure when I solve enterprise A's problem, I'm solving it for B, C, D, and E before they've even had a chance to express it.

s57
00:05:02.680 --> 00:05:13.560
So, there's the two kinds of forward deployment engineer that we have internally, configuring the agent and then making sure all the problems that you interact with the enterprise that that come up in that in in in in the context of conversation

s58
00:05:13.560 --> 00:05:16.040
also get brought back into the product.

s59
00:05:16.040 --> 00:05:17.760
Which brings me to a really important point.

s60
00:05:17.760 --> 00:05:26.960
In fact, it's so important I wish I had a slide for it, but uh at Decagon, forward deployment engineering is identical to product engineering.

s61
00:05:26.960 --> 00:05:40.320
Uh it's the same bar, it's the same reporting structure, uh often like the same team, because the the the the delineation between what is historically forward deployment versus product engineering

s62
00:05:40.320 --> 00:05:42.400
is super super blurred now.

s63
00:05:42.400 --> 00:05:51.440
Uh when I'm speaking with a Fortune 20 and they express a pain point, that is often a product feature that needs to get built and prioritized.

s64
00:05:51.440 --> 00:05:56.920
And so that line between I'm a forward deployed person and I'm a person who works on the product um is gone.

s65
00:05:56.920 --> 00:06:02.040
Uh it's the same it's the same person and and that's represented in our in our in our org chart.

s66
00:06:03.200 --> 00:06:11.720
So, um early on, I mean, Degagon is is an example of sort of canonical hypergrowth.

s67
00:06:11.720 --> 00:06:14.009
A year ago we were at 50 people.

s68
00:06:14.009 --> 00:06:14.040
[snorts]

s69
00:06:14.040 --> 00:06:15.840
Now we're at 500.

s70
00:06:15.840 --> 00:06:19.680
And uh the scale is not slowing down.

s71
00:06:19.680 --> 00:06:26.520
So, actually shameless plug, if you are interested in uh a new role, sunny@degagon.ai is my email.

s72
00:06:26.520 --> 00:06:27.840
Please let me know.

s73
00:06:27.840 --> 00:06:31.000
I'll make sure your your resume {slash} profile gets front of my people.

s74
00:06:31.000 --> 00:06:32.760
Anyway, back to the back to the talk.

s75
00:06:32.760 --> 00:06:37.160
Um so, historically we had agent software engineers and they did it all.

s76
00:06:37.160 --> 00:06:41.919
They did that configuring of that agent brain sitting side by side with our customer.

s77
00:06:41.919 --> 00:06:52.560
Uh this is again things like what is the tonality of the agent, what sort of voice do you want it to have, both literally the voice, but also the the way it speaks.

s78
00:06:52.560 --> 00:06:58.120
Um how do I integrate it into your back-end systems so that it could take action on behalf of the users?

s79
00:06:58.120 --> 00:07:04.880
This can be something simple like I want to reset my password, so the agent needs to have back-end access into your, you know, authentication system.

s80
00:07:04.880 --> 00:07:07.919
And it can be something sort of far more complex than that.

s81
00:07:07.919 --> 00:07:15.280
And they also did some of that like platform work like customer A has this feature request and then building that back into the product.

s82
00:07:15.280 --> 00:07:24.640
Now that we're 500 people, uh we start thinking a lot more about how do we design the Degagon system so that it can scale.

s83
00:07:24.640 --> 00:07:30.400
And effectively we we we broke apart this agent software engineering role into two specialized lanes.

s84
00:07:30.400 --> 00:07:33.320
One is the agent builder and these are like Degagon pros.

s85
00:07:33.320 --> 00:07:37.160
They have a lot of intuition for the various models that power our platform.

s86
00:07:37.160 --> 00:07:40.400
How do you make them work for the use case that the enterprise requires?

s87
00:07:40.400 --> 00:07:47.960
Um largely living within the UI to the extent possible, flagging when things need to go off UI and how do we bring that into the product.

s88
00:07:47.960 --> 00:07:49.840
And then secondly, we have agent software engineers.

s89
00:07:49.840 --> 00:07:54.120
Again, these are the front-line enterprise makes product request.

s90
00:07:54.120 --> 00:07:57.520
Making sure that gets incorporated back into the product.

s91
00:07:58.360 --> 00:08:15.919
And um, this is I think like a super super uh, important insight, uh, which is there is routinely this temptation of Okay, customer A made this request and they're so important to us and they want it done ASAP and maybe I'll just go prompt Codex and Cloud Code to just do it for me.

s92
00:08:15.919 --> 00:08:22.880
But the scarce skill now that AI coding is so good, the scarce skill is actually exercising restraint.

s93
00:08:22.880 --> 00:08:29.480
Uh, and saying, you know, really thinking about how does this going to scale to sort of future customers?

s94
00:08:29.480 --> 00:08:30.760
And part of this is our ethos.

s95
00:08:30.760 --> 00:08:39.520
Like we we build agents to be owned by the customer and so if it turns into a black box of like prompts and patches and that's not good for us or them.

s96
00:08:39.520 --> 00:08:41.360
It's far too brittle.

s97
00:08:41.360 --> 00:08:51.240
Um, but also, uh, when you're a forward deployed person, this is this is kind of uh, this is incumbent upon you to to be exercising this restraint of like, let me not do the easy one-off thing,

s98
00:08:51.240 --> 00:08:58.720
but rather make sure whatever I am building is architected in a way that future customers benefit from.

s99
00:08:58.720 --> 00:09:10.040
So this will come up in in the remainder of my 10 minutes here, which is uh, always thinking about how do I make this one ask benefit the remainder of the customers?

s100
00:09:11.360 --> 00:09:20.440
Um, so I I I put this slide here not to sort of toot our own horn, but to actually talk about what it looks like to achieve success.

s101
00:09:20.440 --> 00:09:31.360
Uh, and in in our case, we've learned like early on when you're scoping the deal, like literally when the very first conversations, you want to figure out ahead of time

s102
00:09:31.360 --> 00:09:33.440
what does success look like for the customer.

s103
00:09:33.440 --> 00:09:38.640
And really narrowing that down, ideally getting it in writing so that there is like no miscommunication along the way.

s104
00:09:38.640 --> 00:09:49.520
Like and when I say what does success look like, I mean what are the metrics you're trying to hit, what sort of channel that you want support on, maybe that's a phone call, maybe that's email, maybe that's text, maybe it's WhatsApp, whatever.

s105
00:09:49.520 --> 00:09:55.960
But really narrowing like what is your pain point, what is the ideal outcome you want, and then we can race to go build that out.

s106
00:09:55.960 --> 00:10:06.120
Um but I think again, back to sort of lessons for forward deployed folks, uh there especially when you're dealing with a large company, there's a temptation to just get started.

s107
00:10:06.120 --> 00:10:14.760
And uh and this is partly a reflection of how AI coding has changed engineering generally, but now there's a lot of effort that has to go up front in requirements gathering,

s108
00:10:14.760 --> 00:10:21.080
making sure you're aligned on what actually has to get built uh before before going to do it.

s109
00:10:22.240 --> 00:10:24.360
Uh this has been a really good learning for us.

s110
00:10:24.360 --> 00:10:36.800
So, uh we try now, given that we have like a a ton of customers across various verticals, we have found it's really helpful to have industry experts that get staffed, the same kind of deal.

s111
00:10:36.800 --> 00:10:48.960
So, if I am working on financial service A, B, and C, when financial service D company comes around, ideally I have a core core group of folks who have experience with those customers

s112
00:10:48.960 --> 00:10:51.080
uh working with this new logo.

s113
00:10:51.080 --> 00:10:53.320
And the idea here is like a lot of that knowledge compounds.

s114
00:10:53.320 --> 00:10:58.880
Like A, you can like speak in the lingo of this customer, and therefore there's a lot more credibility there.

s115
00:10:58.880 --> 00:11:02.160
There's a lot more There's a lot like a much faster ramp up.

s116
00:11:02.160 --> 00:11:02.212
And

s117
00:11:02.212 --> 00:11:02.920
[snorts]

s118
00:11:02.920 --> 00:11:06.960
uh a lot of the agent building, sort of the way you think about success carries over.

s119
00:11:06.960 --> 00:11:16.120
So, uh this has been very helpful for us, and and ultimately it's all about, you know, uh making every deployment uh faster than the last one.

s120
00:11:17.960 --> 00:11:26.640
Um I mentioned earlier that as a forward deployed and and by the way, I say forward deployed engineering, but really it's just like all forms of forward deployment.

s121
00:11:26.640 --> 00:11:35.880
I mentioned earlier that one of the big things you have to do is to always think about how do I solve this customer problem in a way that extends to other customers.

s122
00:11:35.880 --> 00:11:44.320
The other thing that I think is always helpful to keep top of mind is how do I make it so that I'm empowering the rest of the business to solve this problem?

s123
00:11:44.320 --> 00:11:46.440
And this is specifically if you're an engineering.

s124
00:11:46.440 --> 00:11:54.840
So, for example, uh Deckagon's ethos is you should be able to configure this agent completely via natural language.

s125
00:11:54.840 --> 00:12:00.080
And so, if you ever have an engineer needing to do something that needs to get upstreamed back into the product.

s126
00:12:00.080 --> 00:12:07.000
Uh and so, this is sort of a funnel that we have of like, "Look, Deck front forward play engineering, they're the front line for customer asks."

s127
00:12:07.000 --> 00:12:10.920
Um but really it should get it should get sort of scaled across the business.

s128
00:12:10.920 --> 00:12:14.520
And one example of this, and I mentioned it later as well, is like, let's take an integration.

s129
00:12:14.520 --> 00:12:18.480
Let's say Deckagon needs to integrate into like some some CRM.

s130
00:12:18.480 --> 00:12:22.840
Uh early on in our history, we were actually like building custom integrations time and time again.

s131
00:12:22.840 --> 00:12:24.000
And then we thought enough is enough.

s132
00:12:24.000 --> 00:12:27.080
After like the 25th one, we're like, "I don't know how many more are coming up."

s133
00:12:27.080 --> 00:12:29.920
Uh so, let's just like build it a self-serve way.

s134
00:12:29.920 --> 00:12:36.960
And now what took an engineer custom code writing can now be self-served by the customer or built by our agent building team.

s135
00:12:36.960 --> 00:12:40.640
So, it's all about how do you scale the work that you're doing.

s136
00:12:41.520 --> 00:12:51.600
Um also very relevant, depending on the kind of forward deployment work you do, is especially in the enterprise, wanting to prove value as fast as possible.

s137
00:12:51.600 --> 00:12:56.280
For those of you who work especially in the Fortune 500, you're going to get hit with the entire What's the expression?

s138
00:12:56.280 --> 00:12:57.520
Kitchen sink or the entire kitchen?

s139
00:12:57.520 --> 00:12:58.520
Something like this.

s140
00:12:58.520 --> 00:13:02.880
Uh but the idea is how do you prove value as fast as possible?

s141
00:13:02.880 --> 00:13:05.720
So, in our case, Deckagon can become arbitrarily complex.

s142
00:13:05.720 --> 00:13:11.600
You can support all sorts of channels, all sorts of very complex user intents.

s143
00:13:11.600 --> 00:13:21.920
We try to figure out how do we demonstrate value ASAP and not have like a multi-month deal or sorry, multi-month uh time to prove value.

s144
00:13:21.920 --> 00:13:25.360
And once we're there and we're adding value, then we expand, right?

s145
00:13:25.360 --> 00:13:28.240
Because ultimately all of our customers are a multi-year partnership.

s146
00:13:28.240 --> 00:13:42.280
And so, we want to make sure we're we're helping you across your entire support flow and in your revenue generating workflows, but it's important as a forward deployed person to figure out, how do I prove value right away and build your build your motion around that.

s147
00:13:44.200 --> 00:13:53.560
Um So, customers will often come to you come to folks and say, uh I want you to do XYZ.

s148
00:13:53.560 --> 00:13:55.480
And and often they're right.

s149
00:13:55.480 --> 00:14:05.840
But, I think as a forward deployed engineer or forward deployed person of any sort, you're you should treat yourself as an advisor rather than just an executor, right?

s150
00:14:05.840 --> 00:14:06.800
You're you're both.

s151
00:14:06.800 --> 00:14:13.680
So, um you're also on the front line of, "Hey, how do I make AI work for the enterprises?"

s152
00:14:13.680 --> 00:14:18.080
And you have so much knowledge because you're seeing it repeated across every single customer.

s153
00:14:18.080 --> 00:14:32.080
And so, what we do at Akkio gone is we actually ingest your historical support data uh and we tell customers that hey, like if you automate this first or this first, this is where actually you'll see the highest ROI.

s154
00:14:32.080 --> 00:14:35.720
Um and sometimes that's not actually what the customer had reached out about.

s155
00:14:35.720 --> 00:14:43.839
Uh and I imagine there's analogs to this across all sorts of verticals, but it's important to keep in mind that your job isn't just an executor.

s156
00:14:43.839 --> 00:14:47.640
It is of course to be an executor, but it is also to be an advisor.

s157
00:14:47.640 --> 00:14:59.160
Uh and to not underrate the fact that you have this domain expertise by being forward deployed across many companies so that you have this knowledge base that's really valuable for the customer to tap into.

s158
00:15:03.120 --> 00:15:09.280
Every time at Akkio gone, someone has to do something manually, we try to make sure it gets upstream back into the product.

s159
00:15:09.280 --> 00:15:17.200
So, I mentioned the integration earlier, but this is I think a good mental model for folks to have if you're on the front lines.

s160
00:15:17.200 --> 00:15:18.960
How do we smoothen out that path?

s161
00:15:18.960 --> 00:15:20.480
Custom becomes self-serve.

s162
00:15:20.480 --> 00:15:22.280
Custom becomes self-serve.

s163
00:15:22.280 --> 00:15:30.760
Um this has become like a guiding ethos for us uh and I suspect it is the case across every kind of forward deployment motion.

s164
00:15:30.760 --> 00:15:34.840
So, I'd encourage everyone in this audience to to um to to keep this top of mind.

s165
00:15:34.840 --> 00:15:37.200
Like, okay, I'm doing this I'm doing this one-off thing.

s166
00:15:37.200 --> 00:15:41.920
Presumably other people in the company also are bringing it back into the product.

s167
00:15:44.000 --> 00:15:44.520
Okay.

s168
00:15:44.520 --> 00:15:52.920
So, um Decky on is really interesting in that it was started by I think now they're in their early 30s, but it was I think at the time they're in their early 20s.

s169
00:15:52.920 --> 00:15:54.280
Oh, sorry, late 20s.

s170
00:15:54.280 --> 00:16:00.560
Um and so, what did what did we do right uh to to sort of deserve the place that we have?

s171
00:16:00.560 --> 00:16:06.120
And I think one is we're known in the industry to move really really fast on customer asks.

s172
00:16:06.120 --> 00:16:08.760
And part of this is just like it's a very hard-working group of folks.

s173
00:16:08.760 --> 00:16:16.960
Um so, that's like a big reason that we got here uh is that we just move really fast deal by deal.

s174
00:16:16.960 --> 00:16:21.720
Thing number two, we've earned trust with customers that we are advisors, not just executors.

s175
00:16:21.720 --> 00:16:32.480
So, we'll we'll we'll we'll we'll be able to tell you based on what we're seeing across all the customers, based on the data you give us um what is going to be the highest ROI for you.

s176
00:16:32.480 --> 00:16:38.080
And then number three, we've been really good at um making sure we productize custom work.

s177
00:16:38.080 --> 00:16:46.960
But, the way we think about this has changed a lot in the last year because again, a a year ago we were 50 people, could all fit on, you know, a lengthy lunch table.

s178
00:16:46.960 --> 00:16:48.280
And now we're 500.

s179
00:16:48.280 --> 00:16:52.520
So, now we think a lot about designing the system.

s180
00:16:52.520 --> 00:17:03.800
So, every time now now we're very rigorous about sharing knowledge across deployments, but uh making sure you extend uh the field the people in the field feed information back to the platform,

s181
00:17:03.800 --> 00:17:07.760
making sure the agent compounds every single time it interfaces with the customer.

s182
00:17:07.760 --> 00:17:17.439
So, if the agent interfaces with customer A, you improve that for customer B. And it's all about sort of taking knowledge from the field and bringing it back into the product.

s183
00:17:17.920 --> 00:17:23.680
Uh and so, just to wrap up here, uh sort of the a few of the few of the themes.

s184
00:17:23.680 --> 00:17:28.960
Number one, obviously you have to make sure you can figure that agent, do whatever the customer wants.

s185
00:17:28.960 --> 00:17:33.280
But number two, uh make sure that it gets fed back into the product.

s186
00:17:33.280 --> 00:17:35.840
And three, mind that funnel that I mentioned earlier.

s187
00:17:35.840 --> 00:17:44.400
You're on the forward, you're on you're on the field, but you want to make sure it scales and make sure it improves uh every every uh future customer interaction.

s188
00:17:44.400 --> 00:17:46.960
Uh again, my my email is sunny@deciagon.ai.

s189
00:17:46.960 --> 00:17:48.040
I'll also be out here.

s190
00:17:48.040 --> 00:17:49.600
Folks have questions.

s191
00:17:49.600 --> 00:17:53.000
Thank you for coming to talk and I hope this is helpful.

s192
00:18:06.657 --> 00:18:08.657
[music]