1
00:00:00,139 --> 00:00:02,566
Hi friends, and welcome
back to APIs You Won't Hate.

2
00:00:02,624 --> 00:00:06,376
My name is Mike Befolco, and I'm sitting
down today with a new friend to have a

3
00:00:06,378 --> 00:00:08,853
discussion about docs docs of all sorts.

4
00:00:09,106 --> 00:00:12,896
If you're an API developer, you've no
doubt been exposed to many of them, and

5
00:00:12,956 --> 00:00:17,693
I'm really interested to meet and talk
with Faizan Khan from Docsalot where we'll

6
00:00:17,693 --> 00:00:21,172
be talking quite a bit, I'd imagine, about
documentation, but also how the nature of

7
00:00:21,172 --> 00:00:25,781
documentation is changing, the way people
use docs is changing, and to be perfectly

8
00:00:25,782 --> 00:00:28,971
honest with you, the way not people use
docs is changing quite a bit as well.

9
00:00:28,973 --> 00:00:30,549
Faizan, thanks so much for joining today.

10
00:00:30,549 --> 00:00:32,612
I am super psyched to
have you on the show.

11
00:00:32,612 --> 00:00:33,122
How are you doing?

12
00:00:33,784 --> 00:00:34,081
I'm good.

13
00:00:34,083 --> 00:00:35,425
Thanks so much for having me.

14
00:00:35,425 --> 00:00:35,760
Excited for this

15
00:00:35,850 --> 00:00:36,851
Yeah, of course.

16
00:00:36,865 --> 00:00:39,837
Tell me a little bit about yourself
and how you got into the p- the

17
00:00:39,838 --> 00:00:42,815
place where you are right now, and
I wanna talk about Docslot and, I

18
00:00:42,816 --> 00:00:44,199
don't know, the origin story as well.

19
00:00:44,813 --> 00:00:45,387
Yeah, for sure.

20
00:00:45,389 --> 00:00:49,104
So I'm the CEO of Docsalot
as in Sir Docs a Lot.

21
00:00:49,151 --> 00:00:53,988
So we create API docs CLIs,
SDKs taking in OpenAPI specs.

22
00:00:53,991 --> 00:00:55,203
Yeah, a bit about the background.

23
00:00:55,214 --> 00:00:58,274
We started as a dev tool product,
so it was called SlashML.

24
00:00:58,322 --> 00:01:02,335
We went through Alchemist Accelerator,
and we were targeting developers,

25
00:01:02,344 --> 00:01:03,557
so trying to get them onboarded.

26
00:01:03,575 --> 00:01:07,197
And as you do, you start bottom up
you build a lot of docs, a lot of

27
00:01:07,203 --> 00:01:10,696
content, a lot of dev tooling around
that, SDKs and stuff like that.

28
00:01:10,696 --> 00:01:14,688
And through that, we built an internal
tool that was way better than any of

29
00:01:14,688 --> 00:01:17,826
the market out there, and SlashML wasn't
getting any traction, so we were like,

30
00:01:17,826 --> 00:01:19,289
"Okay, let's see if we can sell this."

31
00:01:19,634 --> 00:01:23,187
So we just went around, shopped around
and got immediate traction from that.

32
00:01:23,577 --> 00:01:25,265
And that's how Docsalot came into being.

33
00:01:25,265 --> 00:01:27,709
It's been a year now,
almost, on Docsalot itself.

34
00:01:28,135 --> 00:01:30,140
And yeah, here we are as it is.

35
00:01:30,972 --> 00:01:32,704
Yeah, we are definitely here.

36
00:01:32,705 --> 00:01:36,363
And a- always interested to hear
about a pivot and finding traction in

37
00:01:36,364 --> 00:01:37,944
particular, but why don't we start here?

38
00:01:37,944 --> 00:01:40,400
Tell me a little bit about the
elevator pitch for Docslot.

39
00:01:40,409 --> 00:01:42,610
So if I've never heard of it before,
what does Docslot mean to me?

40
00:01:42,767 --> 00:01:42,971
Yeah.

41
00:01:42,971 --> 00:01:47,229
So Docs-a-Lot essentially we call it
the docs for agents and by agents.

42
00:01:47,230 --> 00:01:52,136
So es- essentially what that means
is the-- it in-- we ingest everything

43
00:01:52,159 --> 00:01:55,794
in your company org, so source
code, Slack communication, internal

44
00:01:55,826 --> 00:01:59,419
communication could be OpenAPI
specs, and then we spit out any

45
00:01:59,478 --> 00:02:01,811
agent-facing or customer-facing content.

46
00:02:01,817 --> 00:02:05,056
So it could be documentation,
SDKs, CLIs, MCP server.

47
00:02:05,456 --> 00:02:09,972
And it's essentially a dev-focused
dev marketing agent that's in

48
00:02:09,973 --> 00:02:11,150
your company working with you.

49
00:02:12,413 --> 00:02:13,382
Oh, really interesting.

50
00:02:13,395 --> 00:02:18,936
And so that's as contrasted to the typical
process or like the maybe old world

51
00:02:18,936 --> 00:02:20,569
process of developing documentation.

52
00:02:20,570 --> 00:02:24,251
What did it look like before Docsalot and
how does that differ from what's here now?

53
00:02:24,756 --> 00:02:28,975
Yeah, so before Doxalot, it would be a
normal hu- a human basically taking in,

54
00:02:29,058 --> 00:02:33,392
reading the product analytics seeing
how the people, how users are using the

55
00:02:33,401 --> 00:02:36,875
product, and then chatting with a bunch
of engineers seeing what features are

56
00:02:36,875 --> 00:02:39,517
released, and then just going up there.

57
00:02:39,611 --> 00:02:43,555
Could be any other product, could be
markdown files, updating the change log,

58
00:02:43,559 --> 00:02:48,399
updating the what features were released,
and also just adding new pages to

59
00:02:48,414 --> 00:02:50,638
onboard more people onto their platform.

60
00:02:51,024 --> 00:02:52,683
Doxalot takes all of that out of the way.

61
00:02:52,718 --> 00:02:56,242
And then on top of that, if you
have SDKs, MCP servers, you have

62
00:02:56,252 --> 00:03:00,906
multiple surfaces that the product
manager or the dev marketing or

63
00:03:00,907 --> 00:03:02,514
the dev rel person is maintaining.

64
00:03:02,547 --> 00:03:04,034
So we just help with that.

65
00:03:04,515 --> 00:03:07,860
Having worked in DevRel for a long time
myself, I'm familiar with the process of

66
00:03:07,860 --> 00:03:12,530
maintaining documentation for one SDK in
one language is a fairly intensive thing.

67
00:03:12,784 --> 00:03:14,957
Usually requires some
subject matter expertise.

68
00:03:15,153 --> 00:03:18,592
It definitely requires some use
of the thing, so testing, like if

69
00:03:18,593 --> 00:03:21,538
I'm gonna deploy docs for my SDK,
I should be sure that it works.

70
00:03:21,883 --> 00:03:23,421
And then it requires
some ongoing maintenance.

71
00:03:23,428 --> 00:03:27,799
As the, as bugs are fixed, features
are added things of that nature go on

72
00:03:27,799 --> 00:03:31,613
within the product organization you need
to keep your documentation up to date.

73
00:03:31,738 --> 00:03:34,739
And so there's a lot of that I think
is interesting and requires nuance.

74
00:03:34,820 --> 00:03:37,320
And I think a lot of people might
bristle at the idea that their

75
00:03:37,321 --> 00:03:40,599
documentation would entirely be
written by agents and maintained by

76
00:03:40,602 --> 00:03:43,187
agents, but can you tell me what that
feels like as the end user of it?

77
00:03:43,238 --> 00:03:45,699
So we are not removing
the human from the loop.

78
00:03:45,745 --> 00:03:47,315
We are just augmenting the human.

79
00:03:47,327 --> 00:03:49,179
We are removing the
grunt work out of there.

80
00:03:49,552 --> 00:03:54,693
So we we bring all of the information
onto a dashboard, and then the agent

81
00:03:54,693 --> 00:03:56,268
sort of drafts the first version.

82
00:03:56,279 --> 00:03:59,253
There is always a human approval
required on the final version and

83
00:03:59,253 --> 00:04:01,153
the tweaking and stuff required.

84
00:04:01,491 --> 00:04:04,997
The reason this is helpful is because
there is nowadays so much shipping

85
00:04:04,998 --> 00:04:09,676
going on with agents small teams are
shipping at the rate of 40, 50 engineers.

86
00:04:09,720 --> 00:04:13,902
And con- contrast to that, the-- there
is still a few-- like the ratio of

87
00:04:13,908 --> 00:04:17,387
engineers to human product managers
is still five to one on the lower

88
00:04:17,389 --> 00:04:19,043
end and could be 10 to one, right?

89
00:04:19,457 --> 00:04:23,143
So the features are being shipped
faster, but the customer-facing

90
00:04:23,148 --> 00:04:24,909
stuff is still lagging behind.

91
00:04:25,336 --> 00:04:26,399
So we augment all of that.

92
00:04:26,405 --> 00:04:28,143
People do have internal
workflows for that.

93
00:04:28,144 --> 00:04:30,396
I'm not saying people can't
build that internally.

94
00:04:30,435 --> 00:04:35,320
There are some effort, but when people
or when teams start doing that, they

95
00:04:35,337 --> 00:04:39,183
realize that it's one of those build
versus buy kind of thing, right?

96
00:04:39,238 --> 00:04:41,838
They start maintaining their own
workflows and then they realize, okay,

97
00:04:41,839 --> 00:04:46,700
we need a more reliable sort of system
for this to create PRs, to surface

98
00:04:46,791 --> 00:04:50,215
relevant customer queries so that we
can write relevant pages for them.

99
00:04:51,186 --> 00:04:52,615
That all sounds very familiar to me.

100
00:04:52,705 --> 00:04:55,599
And I think having an automated
hand in some of that certainly

101
00:04:55,605 --> 00:04:58,200
sounds like it would be helpful
and could save a fair bit of time.

102
00:04:58,767 --> 00:05:01,649
You mentioned earlier that one of the
things that the product does is that

103
00:05:01,649 --> 00:05:04,485
you can integrate it with your team's
Slack, and you might have mentioned

104
00:05:04,503 --> 00:05:05,962
Discord, some other things as well there.

105
00:05:06,186 --> 00:05:08,598
But even-- So I have been using
Slack for quite a while too.

106
00:05:08,610 --> 00:05:11,997
That's an interesting idea to me too
because a lot of the context of what's,

107
00:05:12,020 --> 00:05:15,761
what is happening on building out a
product happens in DMs, it happens in

108
00:05:15,762 --> 00:05:19,271
public channels, it happens in meetings,
it happens in planning tools like

109
00:05:19,280 --> 00:05:20,849
Jira or Linear or whatever you use.

110
00:05:21,039 --> 00:05:21,627
Tell me about that.

111
00:05:21,627 --> 00:05:24,968
Like, how is it listening to Slack, and
w- do you get questions from your users

112
00:05:24,969 --> 00:05:27,900
about "Oh, should I be nervous that I'm
publishing pictures of my cats in Slack

113
00:05:27,902 --> 00:05:29,496
and that Doxolot is listening to that?"

114
00:05:29,818 --> 00:05:30,083
No.

115
00:05:30,087 --> 00:05:34,108
So Slack is like like one of the, the
issue that we ran into is Slack is

116
00:05:34,118 --> 00:05:35,762
similar to any GitHub integration.

117
00:05:35,769 --> 00:05:38,951
Like you go through the IT team or
the VP of engineering, and you're

118
00:05:38,980 --> 00:05:42,362
really specific about the kind of
accesses that you're getting in Slack.

119
00:05:42,607 --> 00:05:45,294
So the way it works is it's just
a bot that you install, and it

120
00:05:45,294 --> 00:05:47,962
only gets invoked if you tag it.

121
00:05:48,356 --> 00:05:50,022
So it's not always on.

122
00:05:50,258 --> 00:05:54,553
In Discord it's always on, but in,
in Discord's case we have-- you would

123
00:05:54,560 --> 00:05:59,050
create a special sort of channel in
which it's always on, which is most of

124
00:05:59,050 --> 00:06:02,466
the time a customer support channel,
so that when a customer starts a

125
00:06:02,467 --> 00:06:06,363
query, the bot responds first, and
then there is a human beneath that.

126
00:06:06,598 --> 00:06:11,099
In Slack's case it's just a bot that
you tag as you would a human coworker.

127
00:06:11,463 --> 00:06:14,646
And then a normal query would be
like, "Go look at the previous

128
00:06:14,649 --> 00:06:17,993
threads or the previous discussion
and just create a page out of this

129
00:06:18,005 --> 00:06:19,707
for me to review later." That's it.

130
00:06:20,006 --> 00:06:21,270
Like that's as far as we go.

131
00:06:21,937 --> 00:06:24,619
for the folks listening, if you've never
had the opportunity to build a Slack

132
00:06:24,624 --> 00:06:28,436
bot, it is a little bit baffling and
intimidating at first, but you will get

133
00:06:28,437 --> 00:06:29,895
that familiar, what would you call it?

134
00:06:29,895 --> 00:06:32,204
Like OAuth consent screen where
it asks you what you wanna give

135
00:06:32,204 --> 00:06:33,422
it access to and all that, and

136
00:06:34,205 --> 00:06:34,542
yeah

137
00:06:34,669 --> 00:06:37,732
your stuff can do is brokered
by Slack's agreements there too.

138
00:06:38,414 --> 00:06:38,762
Yeah.

139
00:06:38,825 --> 00:06:42,862
Yeah, it's also like we go through a
pretty nitty-gritty detail of onboarding.

140
00:06:42,870 --> 00:06:46,465
At this point, like every IT
team has evolved the Slack bot.

141
00:06:46,514 --> 00:06:50,466
So we are pretty, Like we initially
with Slack also started with always on

142
00:06:50,480 --> 00:06:53,583
like Discord, but like you mentioned,
like you get access to everything,

143
00:06:53,593 --> 00:06:55,272
which is not what most teams want

144
00:06:56,137 --> 00:06:56,508
Sure.

145
00:06:56,610 --> 00:06:59,169
Yeah, I think that's the first thing
that I would be a little nervous about.

146
00:06:59,376 --> 00:07:02,853
And genuinely that's because I publish or
pub- publish, I share too many memes with

147
00:07:02,855 --> 00:07:05,919
my team and, those don't need to leak into
our documentation or whatever it may be.

148
00:07:06,354 --> 00:07:08,790
So it sounds like due to the nature
of how the product works, it is

149
00:07:08,949 --> 00:07:12,332
largely a tool for communication and
obviously generating documentation.

150
00:07:12,332 --> 00:07:15,922
Are there any limitations or edges in
terms of compatibility with languages?

151
00:07:15,926 --> 00:07:18,240
Will it work with Rust and Python
and all the things, or is it

152
00:07:18,243 --> 00:07:20,293
- specific to certain I guess stacks?

153
00:07:20,966 --> 00:07:22,503
No, it pretty much works with everything.

154
00:07:22,539 --> 00:07:27,630
The only limitation on is on the SDKs
that we generate from OpenAPI specs.

155
00:07:27,676 --> 00:07:29,929
We only generate three so far.

156
00:07:30,287 --> 00:07:33,794
We can upon request, we can do other
things, but like the three are we are

157
00:07:33,814 --> 00:07:37,024
con-- is what we are confident about,
and you can just get started with it.

158
00:07:37,056 --> 00:07:38,980
Python, TypeScript, and Golang.

159
00:07:39,533 --> 00:07:42,463
The rest, it's a question of like
we haven't received any customer

160
00:07:42,463 --> 00:07:46,167
requests and not enough pull for us
to put much effort on that front.

161
00:07:46,970 --> 00:07:50,322
Among the many things that DocsALot
ingests, then it sounds like OpenAPI is

162
00:07:50,323 --> 00:07:53,570
one of those things, and it'll read your
OpenAPI spec and generate documentation,

163
00:07:53,570 --> 00:07:55,352
yes, but SDKs as well from it.

164
00:07:55,723 --> 00:07:56,565
That's really interesting.

165
00:07:56,570 --> 00:07:59,701
That's like a highly competitive
marketplace for sure, but definitely

166
00:07:59,701 --> 00:08:02,816
something where if it's embedded in
your doc system as well or your process

167
00:08:02,816 --> 00:08:05,986
for generating docs and creating docs,
sounds like it'd be really interesting.

168
00:08:06,286 --> 00:08:09,891
I was looking at your site for
DocsALot, the marketing site, and

169
00:08:09,891 --> 00:08:11,119
going through the list of features.

170
00:08:11,259 --> 00:08:12,189
There's quite a bit there.

171
00:08:12,190 --> 00:08:13,052
It's really interesting.

172
00:08:13,060 --> 00:08:16,811
Just to enumerate a few off the top of my
head, obviously documentation generation.

173
00:08:17,066 --> 00:08:21,518
There is SDKs generation MCPs and
then some benchmarking things,

174
00:08:21,542 --> 00:08:22,970
like lots of stuff going on here.

175
00:08:22,970 --> 00:08:25,384
I'd love to talk about any of that that
you're interested in chatting through.

176
00:08:25,659 --> 00:08:26,474
Yeah, for sure.

177
00:08:26,521 --> 00:08:30,262
we could talk about how sort of the
benchmarking thing or how agents are

178
00:08:30,262 --> 00:08:33,610
rather using the docs and how we are
thinking about that, because that's an

179
00:08:33,611 --> 00:08:39,254
active area of interest among all DevTool
people and in, in how to get more agents

180
00:08:39,255 --> 00:08:41,364
to use their product rather than humans.

181
00:08:41,443 --> 00:08:42,655
So let's start with that then.

182
00:08:42,655 --> 00:08:45,038
What do we, what do you mean agents
are reading my documentation?

183
00:08:45,040 --> 00:08:46,836
What is that actually under the covers?

184
00:08:46,955 --> 00:08:51,031
S-so it's essentially you must have
observed it in your workflow or any

185
00:08:51,032 --> 00:08:55,522
of your engineers' workflow, like all
of us use agents now, like Co-Codex,

186
00:08:55,552 --> 00:08:57,540
Claude, Cursor, it could be anything.

187
00:08:57,600 --> 00:09:03,320
Like we don't-- When we try a new tool,
all we do is ask Codex to "Okay, sign

188
00:09:03,320 --> 00:09:07,740
me up onto that, that tool." So that's
how most of the onboarding is happening

189
00:09:07,742 --> 00:09:11,675
nowadays, which means that agents are your
primary-- If you're a DevTool founder,

190
00:09:11,700 --> 00:09:14,103
agents are your primary customer, right?

191
00:09:14,125 --> 00:09:18,299
Like you want to make sure that the
agents can onboard itself in as few steps

192
00:09:18,300 --> 00:09:23,733
as possible without doing multiple tool
calls without ingesting a lot of content.

193
00:09:23,787 --> 00:09:27,391
It should just be like read one page
and you're there and all of it comes

194
00:09:27,391 --> 00:09:29,448
down to the way agents operate.

195
00:09:29,448 --> 00:09:34,292
Like we have a bunch of-- Like we,
Because like we think about that a lot,

196
00:09:34,428 --> 00:09:38,884
like we observe agentic traffic and
want to make sure that our customers

197
00:09:38,885 --> 00:09:41,362
get more customers or agents rather.

198
00:09:41,373 --> 00:09:45,648
So we-- Like that's one of the key
areas that I'm spending a lot of my time

199
00:09:45,649 --> 00:09:50,225
on, just analyzing a bunch of traffic
that people are getting on their docs.

200
00:09:50,333 --> 00:09:53,893
So one of the things we do, which
is a secret, is that we are on

201
00:09:53,906 --> 00:09:56,767
Cloudflare, so we route all of
the traffic through Cloudflare.

202
00:09:56,974 --> 00:10:01,153
So before Cloudflare even had the
bot or the differentiation between

203
00:10:01,508 --> 00:10:03,363
Codex and Claude it was just bot.

204
00:10:03,522 --> 00:10:08,507
Like you had no idea if it was a crawl-
crawler, Google crawler bot or a spam

205
00:10:08,509 --> 00:10:11,529
bot or a cursor query or a Codex query.

206
00:10:11,925 --> 00:10:15,403
So which means if you had that disabled,
you were not getting any requests.

207
00:10:15,567 --> 00:10:18,448
Like your docs were just
returning a 403 or a 404.

208
00:10:19,623 --> 00:10:20,855
So we let everything in.

209
00:10:21,243 --> 00:10:24,568
Like we were like, okay, let's just see
if we can do some empirical analysis

210
00:10:24,570 --> 00:10:27,591
on that ourselves, and that gives us
some-- gave us some really interesting

211
00:10:27,591 --> 00:10:34,049
insights, like where to place the
index.md or the llms.txt on every page.

212
00:10:34,673 --> 00:10:39,413
Because the way agents read file
is they go they go depth first.

213
00:10:39,470 --> 00:10:40,646
So they don't browse.

214
00:10:40,678 --> 00:10:42,282
They don't look at the index file.

215
00:10:42,305 --> 00:10:47,636
They will just find the nearest matching
page to the query and just read that

216
00:10:47,674 --> 00:10:51,634
page all the way to the end and be like,
"Okay, I'm done now. This is it." The

217
00:10:51,658 --> 00:10:56,609
onboarding page does not let me onboard
Or the onboarding page only has support

218
00:10:56,615 --> 00:10:58,726
for X, Y, Z, and therefore we are done.

219
00:10:58,727 --> 00:11:02,774
Or in, in worst cases early on, it
would also hallucinate the answer.

220
00:11:02,792 --> 00:11:05,855
So it would be like the onboarding
page mentioned this, and that, and

221
00:11:06,032 --> 00:11:09,706
I'm gonna pattern match it against
the other docs that I've read

222
00:11:09,707 --> 00:11:11,692
and do-- figure out what it does.

223
00:11:12,108 --> 00:11:14,346
What you want in that case
is you want it to recover.

224
00:11:14,408 --> 00:11:19,711
You want it to go back to your index
file and expo- explore other pages

225
00:11:20,277 --> 00:11:22,004
where the answer might exist, right?

226
00:11:22,382 --> 00:11:25,736
So that recovery part is really important,
and we experimented with a bunch of

227
00:11:25,737 --> 00:11:28,881
things like putting the LLMs at the
top, putting the LLMs at the bottom,

228
00:11:29,137 --> 00:11:31,659
LLMs or TXT in, in every, every page.

229
00:11:32,038 --> 00:11:33,608
In the middle try to stick in.

230
00:11:33,611 --> 00:11:38,194
And what we found is if you just put it
in the first column and have one simple

231
00:11:38,194 --> 00:11:42,692
line that says, "Hey, if you're an
agent, we have this llms.txt contains a

232
00:11:42,692 --> 00:11:45,010
list of all of the pages in our docs."

233
00:11:45,027 --> 00:11:45,573
Yeah.

234
00:11:45,815 --> 00:11:46,281
Wow.

235
00:11:46,732 --> 00:11:49,132
You've just said something that's
very simple that is a little

236
00:11:49,132 --> 00:11:50,994
bit shocking to me in some ways.

237
00:11:51,015 --> 00:11:52,853
But okay, let me make sure I understand.

238
00:11:52,853 --> 00:11:56,857
So a- agents are coming and browsing
my doc site to learn how to use my

239
00:11:56,857 --> 00:12:00,800
tool because people are sending it to
me from Cursor or Claude or Codex, and

240
00:12:00,812 --> 00:12:04,966
they are using a very algorithm-driven
way of seeing what's on my site.

241
00:12:04,969 --> 00:12:07,911
So when they get there, they find
the way to go to llms.txt, which is a

242
00:12:08,002 --> 00:12:11,902
text file that explains what LLMs have
access to on my site, or I think you

243
00:12:11,903 --> 00:12:14,070
said index.md, which is a similar idea,

244
00:12:14,153 --> 00:12:14,916
both of, yeah

245
00:12:15,387 --> 00:12:15,737
Yeah.

246
00:12:16,185 --> 00:12:19,249
And so what you're saying is it gets
there and it does the first thing it can

247
00:12:19,250 --> 00:12:22,757
find on that first tier is if it says
this is onboarding, it thinks that's the

248
00:12:22,763 --> 00:12:24,130
onboarding, that's the tendency of it.

249
00:12:24,528 --> 00:12:27,707
And so the escape route is to then give
them something that says, "Hey, just

250
00:12:27,707 --> 00:12:30,012
so you know, as you're reading through
this, there's more stuff here. Go here

251
00:12:30,012 --> 00:12:34,503
if you're a bot and get all the stuff."
That is… That's really interesting.

252
00:12:34,505 --> 00:12:35,208
That's very cool.

253
00:12:35,241 --> 00:12:37,099
And obviously not how humans work, right?

254
00:12:37,099 --> 00:12:38,582
People don't tend to
look at things that way.

255
00:12:38,699 --> 00:12:39,070
Yeah.

256
00:12:39,070 --> 00:12:42,369
if you have multiple onboarding
paths, like in our docs, we have

257
00:12:42,370 --> 00:12:46,065
a CLI onboarding and a web editor
onboarding and an MCP onboarding.

258
00:12:46,081 --> 00:12:47,450
Like three types of onboardings.

259
00:12:47,816 --> 00:12:52,431
If you ask the agent to go onboard
itself to Docs a lot, it's if we didn't

260
00:12:52,442 --> 00:12:56,202
have the LMS or DXT, it's just gonna
be like, "Okay they only have web

261
00:12:56,203 --> 00:13:00,522
editor onboarding, so therefore I'm
gonna open the browser in Codex and try

262
00:13:00,522 --> 00:13:02,484
to sign up via your Google account."

263
00:13:02,502 --> 00:13:07,215
But in that case, what we want is for
for it to use the CLI or the MCP, right?

264
00:13:07,985 --> 00:13:10,212
Which is… So yeah, in that
case it's really helpful.

265
00:13:10,908 --> 00:13:11,306
Yeah.

266
00:13:11,369 --> 00:13:12,381
Oh, that's really interesting.

267
00:13:12,709 --> 00:13:16,702
Okay, so pedaling back a little bit
then, writing docs for agents and for

268
00:13:16,703 --> 00:13:20,733
people, the, the actual, meat and potatoes
of the docs of how to do things like,

269
00:13:20,759 --> 00:13:22,960
let's say, the "Hello, world" of my SDK.

270
00:13:23,108 --> 00:13:26,018
Is there a difference in writing that for
people versus writing that for an agent?

271
00:13:26,068 --> 00:13:28,541
Or is it more a matter of how you
tell it to get to other things?

272
00:13:29,662 --> 00:13:32,748
I think it's subtle differences
we don't have that much… so

273
00:13:32,790 --> 00:13:35,022
agents want negation on every page.

274
00:13:35,127 --> 00:13:37,963
If you don't have a negation like this
is what this page does and this is

275
00:13:37,964 --> 00:13:42,318
what this page does not do the agent is
going to be like, if you remo- if you

276
00:13:42,322 --> 00:13:46,463
leave out the do not do part, the agent
is gonna be like, "This is the page."

277
00:13:46,550 --> 00:13:47,032
This is it.

278
00:13:47,413 --> 00:13:51,225
And there- therefore, because it
pattern matches it against text, right?

279
00:13:51,291 --> 00:13:53,986
So it's if you say let's
say the onboarding.

280
00:13:54,055 --> 00:13:58,537
If your onboarding page does not
explicitly say that this is web editor

281
00:13:58,552 --> 00:14:02,542
only onboarding and we have other
onboarding parts as well, it's gonna

282
00:14:02,549 --> 00:14:05,080
be like, "Okay, this is the only
onboarding that I could find, and

283
00:14:05,113 --> 00:14:06,451
therefore I'm just gonna use this."

284
00:14:06,877 --> 00:14:08,679
Whereas humans will explore.

285
00:14:08,707 --> 00:14:11,739
Humans will be like, "Okay, let me
see. Let me click a few buttons,"

286
00:14:11,852 --> 00:14:12,287
r-right.

287
00:14:12,340 --> 00:14:12,650
Yeah.

288
00:14:12,650 --> 00:14:17,203
And to use a weird parallel, humans'
context can go beyond that session.

289
00:14:17,543 --> 00:14:20,102
I may have been on this page yesterday
and remember that there's an MCP

290
00:14:20,110 --> 00:14:21,227
or something that I can use here.

291
00:14:21,773 --> 00:14:24,157
Oh, man, that-- what a really
interesting space to be in.

292
00:14:24,167 --> 00:14:27,186
That is at the edge of a lot of
things are changing very quickly

293
00:14:27,186 --> 00:14:28,645
and how do we all do this well?

294
00:14:28,985 --> 00:14:33,111
And so your you said you found a lot of
traction with this product pretty quickly.

295
00:14:33,299 --> 00:14:34,783
What did your first customers look like?

296
00:14:34,784 --> 00:14:37,044
The people that were actually
adopting it, what were-- did

297
00:14:37,047 --> 00:14:37,875
they have things in common?

298
00:14:37,875 --> 00:14:39,582
Where did they come
from surprising places?

299
00:14:40,269 --> 00:14:40,944
Yeah, mostly.

300
00:14:40,948 --> 00:14:43,181
So like I mentioned, we went
through Alchemist Accelerator and

301
00:14:43,191 --> 00:14:46,637
the early customers were mostly
just us going to them and be like,

302
00:14:46,637 --> 00:14:49,514
"Can we create docs for you?" Like
manually before we had a product.

303
00:14:50,918 --> 00:14:54,410
And that was a clear signal for
Paul "Okay, like we don't have

304
00:14:54,415 --> 00:14:57,359
time. We have a demo day coming
up," and yeah, just do that.

305
00:14:57,703 --> 00:15:02,755
Like we are paying GitBooks and Mandify
who are the sort of the bigger players.

306
00:15:02,788 --> 00:15:05,975
Like we're paying them anyways if you
can do that half the cost without me

307
00:15:06,006 --> 00:15:07,698
touching anything, like just do that.

308
00:15:08,227 --> 00:15:11,225
And at that point, we were of
course using like coding agents

309
00:15:11,228 --> 00:15:12,675
to build all of this ourselves.

310
00:15:12,675 --> 00:15:17,137
And what we initially started with was
like, okay, let's just focus on this.

311
00:15:17,191 --> 00:15:22,118
Le- let's say if you have a Docusaurus
of your docs, can we automate maintaining

312
00:15:22,121 --> 00:15:26,664
Docusaurus, like all of the open
source, like Astro, Starlight Fuma docs?

313
00:15:26,764 --> 00:15:28,673
Like we are platform agnostic.

314
00:15:28,731 --> 00:15:29,859
Like we just maintain the docs.

315
00:15:29,870 --> 00:15:32,524
So we were at the GitHub
level, if you will.

316
00:15:32,891 --> 00:15:36,328
Like we had a Git- GitHub app and
we would ingest source code and then

317
00:15:36,328 --> 00:15:40,658
just spit out changes to the docs in
Markdown or could be HTML file even.

318
00:15:40,999 --> 00:15:44,288
Some people had their own custom
Next.js file, like in Nextra.

319
00:15:45,069 --> 00:15:48,820
And then over time as, as Claude
were, was becoming more mature,

320
00:15:48,838 --> 00:15:51,879
we realized that, okay, people can
build these triggers themselves,

321
00:15:52,458 --> 00:15:53,783
maintain these things themselves.

322
00:15:53,785 --> 00:15:57,347
We just started improving the platform,
improving the design of our docs,

323
00:15:57,365 --> 00:16:02,303
because that's-- when all is said and
done, people still want their docs

324
00:16:02,321 --> 00:16:06,813
to look really good, regardless of
whether agents are using them or not,

325
00:16:06,813 --> 00:16:10,449
because it's it's an expression of who
you are as a company or as a product.

326
00:16:10,846 --> 00:16:12,159
Especially if you're in dev tools.

327
00:16:12,195 --> 00:16:17,923
Developers tend to be really judgy so if
your docs look wipe coded or half done

328
00:16:17,953 --> 00:16:21,638
or not aesthetically pleasing they'll
be like, "Okay, the product is not

329
00:16:22,000 --> 00:16:23,449
that mature." Like they're a bit early

330
00:16:23,679 --> 00:16:26,036
we are our own harshest critics without a

331
00:16:26,068 --> 00:16:26,146
Yeah.

332
00:16:26,750 --> 00:16:27,052
Yeah.

333
00:16:27,456 --> 00:16:27,736
Yeah.

334
00:16:27,758 --> 00:16:33,231
So it doesn't matter if agents are the
users of docs, like there is still a human

335
00:16:33,276 --> 00:16:34,998
at the end of the day making the decision.

336
00:16:35,522 --> 00:16:37,030
So the docs have to look pretty.

337
00:16:37,042 --> 00:16:41,471
Like that's one of the key like rule
for any docs platform or even if

338
00:16:41,473 --> 00:16:44,976
anyone is building it on their own,
like your docs have to look pretty.

339
00:16:44,978 --> 00:16:48,251
Like I've seen a clear differentiation
of in our journey, but also

340
00:16:48,254 --> 00:16:49,292
in our customers' journey.

341
00:16:49,293 --> 00:16:55,006
When you flip the design of the docs, like
you immediately-- people start immediately

342
00:16:55,006 --> 00:16:56,708
seeing you as a more mature product.

343
00:16:57,185 --> 00:16:57,949
The product is the same.

344
00:16:57,951 --> 00:17:01,238
You just change the design, but it's
this like contrasting difference.

345
00:17:01,895 --> 00:17:04,931
And that in itself is a
very human benchmark, right?

346
00:17:04,997 --> 00:17:09,332
If you send an agent to the same docs with
two flavors of theme or however you wanna

347
00:17:09,332 --> 00:17:13,564
look at it, they don't care, They wanna
know if it works, and maybe that's more

348
00:17:13,564 --> 00:17:17,020
objective certainly, but there's still
lots of humans making decisions here and

349
00:17:17,021 --> 00:17:18,589
actually, often swiping the credit card.

350
00:17:18,655 --> 00:17:19,780
That's an important angle there.

351
00:17:20,196 --> 00:17:22,319
I would love to talk a little
bit about how this is all built.

352
00:17:22,346 --> 00:17:25,382
You've mentioned a few bits of
tech here, so some Cloudflare

353
00:17:25,390 --> 00:17:27,129
stuff and a few different LLM tech.

354
00:17:27,130 --> 00:17:28,708
Like what, what's under the covers?

355
00:17:29,582 --> 00:17:29,986
Yeah.

356
00:17:29,996 --> 00:17:33,920
So it's a-- So the dashboard
is built in Next.js platform.

357
00:17:33,982 --> 00:17:37,982
We are on Google and Cloudflare
in the middle for load

358
00:17:37,982 --> 00:17:38,926
balancing and other things.

359
00:17:38,928 --> 00:17:43,501
And then for every sort of customer docs,
we spin up a separate Cloud Run instance.

360
00:17:43,912 --> 00:17:47,509
That helps us make things
really isolated and simple.

361
00:17:47,536 --> 00:17:50,735
And in case of if it's an enterprise
customer, all we do is just increase

362
00:17:50,738 --> 00:17:52,487
the number of load balancing.

363
00:17:52,862 --> 00:17:56,782
And if it's a small sort of early
pro-tier customer we just spin

364
00:17:56,783 --> 00:17:57,999
up a simple smaller instance.

365
00:17:57,999 --> 00:18:02,603
It makes it easy for me to price the
product and for customers… and it

366
00:18:02,653 --> 00:18:05,823
also makes it just more secure, like
everything is sandboxed and isolated.

367
00:18:05,873 --> 00:18:08,225
Like no, no environment
leakage is happening.

368
00:18:08,276 --> 00:18:09,430
We don't have one box.

369
00:18:09,526 --> 00:18:13,210
Yeah, and that was intentional by
design when we were building the

370
00:18:13,210 --> 00:18:14,775
platform around the same time.

371
00:18:14,921 --> 00:18:18,398
I don't want to name them, but like
one of the bigger platform had a

372
00:18:18,399 --> 00:18:23,504
similar hack happened, and that was
one of the hack is someone got access

373
00:18:23,504 --> 00:18:27,484
to one box and information leakage
did happen from other customers.

374
00:18:27,518 --> 00:18:29,861
So we were like really particular
about okay, let's delay the

375
00:18:29,861 --> 00:18:32,881
launch, but let's make sure we
have separate boxes for everything.

376
00:18:33,494 --> 00:18:33,825
Yeah.

377
00:18:34,008 --> 00:18:37,833
I guess probably in your case too, that
necessarily means that your, like the

378
00:18:37,833 --> 00:18:41,706
bot that has context about my product
only can get context about my product

379
00:18:41,738 --> 00:18:44,631
'cause it's cordoned off to that
thing and, is listening to my team by

380
00:18:44,755 --> 00:18:46,268
whatever ways too, I would imagine.

381
00:18:46,763 --> 00:18:49,208
Yeah, that one yeah, that one as well.

382
00:18:49,280 --> 00:18:55,497
But the-- it's also more around like the
if you have a GitHub app installed in your

383
00:18:55,534 --> 00:19:00,722
organization and if someone gets access to
that instance the- they don't have access

384
00:19:00,723 --> 00:19:02,983
to other other instances basically, right?

385
00:19:02,988 --> 00:19:04,428
So other organizations, right?

386
00:19:04,783 --> 00:19:08,074
Hacks can happen, like you can't defend
against them, but like it's easier for

387
00:19:08,074 --> 00:19:12,958
us to just spin down one instance versus
one box that has 1,000 customers, right?

388
00:19:13,093 --> 00:19:14,557
Yeah, that could be problematic,

389
00:19:14,900 --> 00:19:15,180
Yeah.

390
00:19:15,850 --> 00:19:16,228
Yeah.

391
00:19:16,273 --> 00:19:16,590
Okay.

392
00:19:16,592 --> 00:19:19,342
You mentioned a minute ago as well
predictable pricing for your customers.

393
00:19:19,379 --> 00:19:20,816
Let's talk a little bit
about the pricing model.

394
00:19:20,818 --> 00:19:21,964
What does it cost to get onboarded?

395
00:19:21,977 --> 00:19:24,174
What is in terms of dollars but also time?

396
00:19:24,204 --> 00:19:26,361
What does it take for people to get
started, and what's it gonna cost them

397
00:19:26,361 --> 00:19:27,976
the first time they spin up some docs?

398
00:19:28,557 --> 00:19:28,748
Yeah.

399
00:19:28,798 --> 00:19:32,062
S-so one caveat is like we are
not-- we don't have a free tier.

400
00:19:32,159 --> 00:19:37,561
The if people want a free service,
they can go to one of the competitors,

401
00:19:37,561 --> 00:19:40,036
just search for them, and then they
can spin up a good-looking docs.

402
00:19:40,375 --> 00:19:44,540
We are… So our lower tier
starts at 39, and then it's 99 for

403
00:19:44,541 --> 00:19:45,898
the pro tier, which gets you…

404
00:19:46,349 --> 00:19:48,995
And the difference between the two is
just if your docs are receiving more

405
00:19:48,997 --> 00:19:52,655
traffic and you want more automations
you just move to the higher tier.

406
00:19:52,665 --> 00:19:56,958
And then we have the enterprise
tiers, which has SSO and other things.

407
00:19:56,984 --> 00:20:00,152
And it's similar for MCP servers
and SDK generation as well.

408
00:20:00,184 --> 00:20:03,731
If you are doing more updates you
will go to the upper tier, otherwise

409
00:20:03,849 --> 00:20:05,013
on the lower tier it's free.

410
00:20:05,399 --> 00:20:09,295
There is a two weeks free trial, so it's
literally free to try it out, everything.

411
00:20:09,678 --> 00:20:12,259
And then other than that, we
are free core open source.

412
00:20:12,319 --> 00:20:15,493
So we have a bunch of open source
products using the pro tier.

413
00:20:15,866 --> 00:20:18,740
And yeah, so it's commercial and
non-commercial, so we give it away

414
00:20:18,741 --> 00:20:20,460
for free to open source products.

415
00:20:20,939 --> 00:20:24,648
And in terms of how much it
costs us, like the docs platform

416
00:20:24,708 --> 00:20:25,786
is really simple to price.

417
00:20:25,821 --> 00:20:28,621
Essentially, if you remove
all of the AI automation.

418
00:20:28,656 --> 00:20:30,327
AI is a bit harder to price.

419
00:20:30,389 --> 00:20:34,465
Like you don't know the token pricing and
how much how much you are spending versus

420
00:20:34,504 --> 00:20:36,079
the versus how much people are using.

421
00:20:36,120 --> 00:20:38,027
So we are still on a predictable pricing.

422
00:20:38,038 --> 00:20:41,493
Some-- I see some of the people
have gone into credit-based model,

423
00:20:42,017 --> 00:20:46,136
which I feel like makes sense for
enterprise customers, but for a

424
00:20:46,142 --> 00:20:48,700
startup that can be really detrimental.

425
00:20:48,753 --> 00:20:52,365
If they, one of their developer
comes in and makes 500

426
00:20:52,365 --> 00:20:55,057
changes, do they pay 5,000 now?

427
00:20:55,354 --> 00:20:57,176
Like right.

428
00:20:57,853 --> 00:21:00,272
Yeah, it can be unpredictable,
especially with startups when

429
00:21:00,310 --> 00:21:01,516
things are moving really fast.

430
00:21:01,528 --> 00:21:06,611
The CEO does not have time to add all of
the thresholds and automations, right?

431
00:21:06,985 --> 00:21:11,194
And our primary like target
audience is like startups.

432
00:21:11,208 --> 00:21:14,671
Like we are not going after enterprises
like that's how we got most of the pull.

433
00:21:14,671 --> 00:21:17,731
And to be honest, we can't serve
enterprises at our scale as well.

434
00:21:17,765 --> 00:21:20,590
So it's like we love startups,
we are at startups level.

435
00:21:21,029 --> 00:21:21,713
We still have a…

436
00:21:21,745 --> 00:21:22,374
call to action.

437
00:21:22,834 --> 00:21:25,785
If you're build- building an open source
product or running a startup and you

438
00:21:25,786 --> 00:21:30,019
need docs gen of some sort or SDK gen
or MCP gen, like you should be… Your

439
00:21:30,023 --> 00:21:31,314
ears should be perking up right now.

440
00:21:32,269 --> 00:21:32,548
Yeah.

441
00:21:32,592 --> 00:21:32,842
Yeah.

442
00:21:32,894 --> 00:21:37,329
Like our thing is like we are serving
the one, one poor product manager

443
00:21:37,329 --> 00:21:41,573
or the one poor DevRel guy who is
getting bombarded by the features

444
00:21:41,573 --> 00:21:44,872
that their team is shipping and have
no time to just maintain everything.

445
00:21:45,419 --> 00:21:46,735
So that's the person
that we are targeting.

446
00:21:47,509 --> 00:21:49,059
hugely beneficial if you're that guy.

447
00:21:49,097 --> 00:21:51,033
This is the best day of
your life, certainly.

448
00:21:51,601 --> 00:21:55,747
We were also chatting before I hit record
about something that we're, we are happen

449
00:21:55,748 --> 00:21:59,102
to be kindred spirits in this world of
of building things for people on the web.

450
00:21:59,149 --> 00:22:00,709
Tell me about All Them APIs.

451
00:22:00,731 --> 00:22:01,236
What's that?

452
00:22:01,706 --> 00:22:06,033
Yeah, so all the APIs started
with just our sort of problem of,

453
00:22:06,095 --> 00:22:09,789
okay, how are agents using the
products, docs, and everything?

454
00:22:09,825 --> 00:22:14,699
And then so all the API is essentially
a set of benchmarks where what all we

455
00:22:14,699 --> 00:22:20,135
do is take public docs of any product,
not just ours, could be any product.

456
00:22:20,228 --> 00:22:24,173
Like recent, we did a cluster analysis
of emails, and then we give it to

457
00:22:24,176 --> 00:22:28,852
Claude and Codex, and then we just
give it one prompt, and the prompt is

458
00:22:28,865 --> 00:22:30,610
the same for all of the benchmarks.

459
00:22:30,651 --> 00:22:33,672
And we're like, "Okay, go onboard
yourself onto this product.

460
00:22:34,455 --> 00:22:38,411
Create any emails, create any accounts
that you want, use any tool that you

461
00:22:38,419 --> 00:22:44,625
have access to, but do not stop un- until
you're done, and log everything." So we

462
00:22:44,627 --> 00:22:48,791
log everything in the process of that
onboarding process, and that gave us a

463
00:22:48,792 --> 00:22:54,008
really interesting sort of ways of how
the Codex is using the tool, but also

464
00:22:54,151 --> 00:22:59,887
what docs and products are more prone
to agent onboarding versus what are not.

465
00:22:59,937 --> 00:23:04,635
Clear limitations that I'll just name
a few if you have them, and if you're

466
00:23:04,635 --> 00:23:08,227
a de- dev tool founder might want
to consider thinking around them.

467
00:23:08,667 --> 00:23:13,429
One of the ones is so if you have a
credit card like trial might want to

468
00:23:13,430 --> 00:23:16,784
consider just adding threshold and
removing the credit card requirement.

469
00:23:17,387 --> 00:23:20,566
Like that's where a lot of like
Firecrawl, I think, was one, and

470
00:23:20,567 --> 00:23:24,026
then a few others, which makes
sense, like you want to remove spam.

471
00:23:24,528 --> 00:23:27,479
But because agents are trying products,
they don't have access to credit

472
00:23:27,479 --> 00:23:32,057
cards, like you want to give them
free stuff to try before, before then.

473
00:23:32,068 --> 00:23:35,746
The other one is which is-- which
could be a bit harder, but so the

474
00:23:35,804 --> 00:23:39,046
the Claude and Codex did have access
to Gmail, like my Gmail account

475
00:23:39,100 --> 00:23:40,892
or a test account that we created.

476
00:23:41,386 --> 00:23:46,113
But if it requires registration, if
there is no social auth, then it just

477
00:23:46,114 --> 00:23:50,069
increases the number of steps, and
it was prone to fail in one of those

478
00:23:50,069 --> 00:23:54,419
steps because the email arrived late,
the two-factor auth arrived late,

479
00:23:54,466 --> 00:23:55,922
and all of those kinds of things.

480
00:23:56,559 --> 00:24:03,747
And then i- in one of the cases which
I won't name, the like if you have

481
00:24:03,751 --> 00:24:08,353
three or four products and if you
just have one docs for that product

482
00:24:08,624 --> 00:24:11,426
again, it, it brings, comes back to the
other things that we were discussing.

483
00:24:11,435 --> 00:24:15,782
Make sure you explicitly say that the
product that you are looking for is that

484
00:24:15,812 --> 00:24:18,237
one not the first page that you found

485
00:24:18,536 --> 00:24:18,992
yeah.

486
00:24:19,617 --> 00:24:22,359
Yeah, that I think you called it before
n- the show the negatives or show

487
00:24:22,360 --> 00:24:23,543
the, show what's not there on that

488
00:24:23,654 --> 00:24:23,936
Yeah.

489
00:24:23,995 --> 00:24:24,005
Yeah.

490
00:24:24,876 --> 00:24:25,174
yeah.

491
00:24:25,188 --> 00:24:26,248
Oh, that's really interesting.

492
00:24:26,500 --> 00:24:29,659
As I've spent a lot of my career thinking
about information architecture and

493
00:24:29,659 --> 00:24:33,698
design of things for people but it really
is a very important difference there.

494
00:24:33,709 --> 00:24:36,276
I feel like I'm furiously taking
notes here and marking down

495
00:24:36,277 --> 00:24:38,824
things mentally to go try out
after we get off the phone here.

496
00:24:39,085 --> 00:24:39,847
That's really interesting.

497
00:24:39,847 --> 00:24:42,671
And so all them APIs is
something, it's public, right?

498
00:24:42,671 --> 00:24:43,517
People can access this.

499
00:24:43,517 --> 00:24:44,433
What's the URL for that?

500
00:24:45,021 --> 00:24:47,199
It's allthemapis.com.

501
00:24:47,258 --> 00:24:50,191
A-L-L T-H-E-M apis.com.

502
00:24:50,969 --> 00:24:51,680
Very straightforward.

503
00:24:51,688 --> 00:24:52,835
I'll make sure that's in the show notes.

504
00:24:52,835 --> 00:24:56,342
And so maybe if you're building out
an SDK of some sort good to give yours

505
00:24:56,343 --> 00:24:59,473
a run through the run-- Yeah, a run
through all their APIs to test it out.

506
00:24:59,965 --> 00:25:01,068
Man, that's really fascinating.

507
00:25:01,153 --> 00:25:03,753
And so you've certainly taken away
a lot from that and it's like a

508
00:25:03,754 --> 00:25:06,660
kindred product that's making your
features better without a doubt.

509
00:25:07,755 --> 00:25:08,820
I'm curious what's next for you.

510
00:25:08,824 --> 00:25:10,920
So what are you excited about
that your team is building

511
00:25:10,921 --> 00:25:12,081
and what's around the corner?

512
00:25:13,053 --> 00:25:15,791
Yeah, so we have lots
of exciting features.

513
00:25:15,814 --> 00:25:20,398
One of the things that we are thinking
on, like I mentioned, is is how can we

514
00:25:20,398 --> 00:25:25,784
get more customers or agents for, for our
customers that, that's using our platform.

515
00:25:26,236 --> 00:25:29,390
So we are thinking of like there
is a bunch of things because

516
00:25:29,405 --> 00:25:30,749
docs is just one thing, right?

517
00:25:30,765 --> 00:25:34,446
Like there are, there is if you have
an MCP server, there is the MCP,

518
00:25:34,446 --> 00:25:39,180
but also all them APIs is also a, a
sort of a, a path in that direction.

519
00:25:39,239 --> 00:25:43,469
Like if you have an API product how
can you get more eyeballs on that?

520
00:25:43,526 --> 00:25:46,728
So one of the things that we
are thinking of is what would

521
00:25:46,731 --> 00:25:48,482
a dev marketing person do?

522
00:25:48,639 --> 00:25:49,019
Can we

523
00:25:49,071 --> 00:25:49,464
oh yeah

524
00:25:49,579 --> 00:25:49,943
that?

525
00:25:50,117 --> 00:25:52,847
How do you get more eyeballs?

526
00:25:52,862 --> 00:25:57,556
Some content generation social media
automation a change log announcement.

527
00:25:57,632 --> 00:25:59,932
How would a human announce that kind of.

528
00:26:00,058 --> 00:26:02,872
So there is some exciting things
coming up in that sense, which will

529
00:26:03,357 --> 00:26:05,865
give our customers more eyeballs and

530
00:26:05,916 --> 00:26:06,685
that's a great idea.

531
00:26:07,694 --> 00:26:08,081
Sure.

532
00:26:08,136 --> 00:26:10,377
That's one of those rituals that
has to happen with every release

533
00:26:10,377 --> 00:26:14,603
and definitely requires human power
to do and hours of their lives.

534
00:26:14,633 --> 00:26:17,199
And there's lots of good things
you can do around branding and

535
00:26:17,271 --> 00:26:18,433
brand voice and all that there.

536
00:26:18,466 --> 00:26:20,058
That's a super cool idea.

537
00:26:20,116 --> 00:26:20,889
I'm really into that.

538
00:26:21,477 --> 00:26:21,793
Okay.

539
00:26:21,835 --> 00:26:25,773
So Farzan, be- before I let you go I'm
a-- I've got two more questions for you.

540
00:26:25,828 --> 00:26:27,166
We we've talked a lot about DocsALot.

541
00:26:27,181 --> 00:26:29,820
I don't think I've said
the URL, docsalot.dev.

542
00:26:29,883 --> 00:26:31,916
There will also be a link in
the show notes to that for folks

543
00:26:31,917 --> 00:26:32,965
to make sure they check it out.

544
00:26:33,271 --> 00:26:36,739
If someone is interested in trying
DocsALot or has feedback for you, where's

545
00:26:36,739 --> 00:26:37,858
the best place to find you online?

546
00:26:38,683 --> 00:26:43,411
So I'm on Twitter,
F-A-I-Z-A-N-1-0-double one four.

547
00:26:43,466 --> 00:26:44,300
We'll link that.

548
00:26:44,324 --> 00:26:47,016
And then on LinkedIn, they can
just find me, Faizan Docsalot.

549
00:26:47,444 --> 00:26:48,353
I should pop up.

550
00:26:48,736 --> 00:26:49,667
Blue Sky as well.

551
00:26:49,723 --> 00:26:53,346
Looks like everywhere, but also
they can just if you log into

552
00:26:53,365 --> 00:26:56,455
any of the websites, you'll see a
support or anything, and I should

553
00:26:56,456 --> 00:26:57,989
pop up on the website at some point.

554
00:26:58,630 --> 00:27:00,292
If I don't, then we are
not doing a good job.

555
00:27:01,376 --> 00:27:02,022
That's perfect.

556
00:27:02,386 --> 00:27:02,836
Sounds good.

557
00:27:02,836 --> 00:27:05,450
Faizan Khan, th- thank you so much
for taking the time to chat with me.

558
00:27:05,452 --> 00:27:06,958
It's been really cool
learning about Docsalot.

559
00:27:06,988 --> 00:27:10,351
I'm excited to see where you'll go, and
please feel free to join us back on the

560
00:27:10,351 --> 00:27:13,414
pod anytime if you've got additional
releases to talk through, or if you

561
00:27:13,414 --> 00:27:15,133
have API chat you wanna get into.

562
00:27:15,865 --> 00:27:16,246
Yeah, for sure.

563
00:27:16,246 --> 00:27:16,955
It's been a pleasure.

564
00:27:16,994 --> 00:27:17,719
We'll do another one.