WEBVTT

1
00:00:13.000 --> 00:00:16.079
<v Speaker 1>Hello everyone, Thanks for joining us from where where you are.

2
00:00:16.879 --> 00:00:20.600
<v Speaker 1>Welcome to another new episode of Startup Still Say Today.

3
00:00:20.600 --> 00:00:25.320
<v Speaker 1>I'm super excited to welcome Vishnu Subra Money into this podcast.

4
00:00:25.760 --> 00:00:28.440
<v Speaker 1>Vishno and I have been talking on and off the

5
00:00:28.559 --> 00:00:32.079
<v Speaker 1>last three months or so. We first met on LinkedIn

6
00:00:32.200 --> 00:00:36.159
<v Speaker 1>through a common post, like many connections these days, and

7
00:00:36.920 --> 00:00:40.000
<v Speaker 1>I also had a great opportunity to meet him in

8
00:00:40.039 --> 00:00:43.679
<v Speaker 1>person while I was visiting India a couple of months ago.

9
00:00:43.799 --> 00:00:46.640
<v Speaker 1>So Vishnu was super kind enough to invite us to

10
00:00:47.399 --> 00:00:50.920
<v Speaker 1>his offices, his studio, his lab which he'll tell us

11
00:00:50.960 --> 00:00:55.399
<v Speaker 1>all about very soon and talk about his journey. And

12
00:00:55.560 --> 00:01:00.200
<v Speaker 1>I thought it was fascinating in this current world and

13
00:01:01.439 --> 00:01:04.840
<v Speaker 1>that we live in where anything and everything we turn on,

14
00:01:05.159 --> 00:01:08.879
<v Speaker 1>whether it's news or a podcast or whatnot, everybody talks

15
00:01:08.879 --> 00:01:12.120
<v Speaker 1>about AI. And we thought, Okay, why don't we bring

16
00:01:12.159 --> 00:01:18.840
<v Speaker 1>Invisioner here and talk here from him about what he

17
00:01:19.000 --> 00:01:23.319
<v Speaker 1>is doing in the world of AI and Soshioner. Welcome

18
00:01:23.359 --> 00:01:26.040
<v Speaker 1>to the show. Hi untanny, thanks for inviting me to

19
00:01:26.079 --> 00:01:29.079
<v Speaker 1>your podcast. I hope you enjoy the session.

20
00:01:29.319 --> 00:01:33.840
<v Speaker 2>A little intro of me. My father was a banker

21
00:01:34.159 --> 00:01:37.680
<v Speaker 2>and I grew up when the banks were getting digitized,

22
00:01:38.280 --> 00:01:41.359
<v Speaker 2>so he had a Unix book. That's how my introdiction

23
00:01:41.439 --> 00:01:44.560
<v Speaker 2>to computers got started. I kind of went through some

24
00:01:44.640 --> 00:01:48.079
<v Speaker 2>of the Linux commands without really practicing. And I've been

25
00:01:48.120 --> 00:01:51.000
<v Speaker 2>fighting at my father from probably my third standard to

26
00:01:51.040 --> 00:01:54.319
<v Speaker 2>get a computer. And I got one when I finished

27
00:01:54.359 --> 00:01:58.040
<v Speaker 2>twelve standard. And I spent most of my colleges in

28
00:01:58.079 --> 00:02:01.879
<v Speaker 2>school bunking them and spending in the college library or

29
00:02:02.000 --> 00:02:04.560
<v Speaker 2>in the browsing center on the gaming center. That's where

30
00:02:04.560 --> 00:02:07.000
<v Speaker 2>I learned a lot. So I've been in the evolution

31
00:02:07.159 --> 00:02:09.840
<v Speaker 2>of a computers much before I understood a lot of things,

32
00:02:10.479 --> 00:02:13.080
<v Speaker 2>like I was an early adopter for Yahoo, Male MS

33
00:02:13.080 --> 00:02:15.800
<v Speaker 2>and hot Mail. Many people may not understand it at

34
00:02:15.800 --> 00:02:18.759
<v Speaker 2>this point of time, but first will understand my first

35
00:02:18.759 --> 00:02:23.400
<v Speaker 2>Internet connection was a personal dialup and things are super slow.

36
00:02:24.280 --> 00:02:27.680
<v Speaker 2>From fighting with my father for getting a computer. Now

37
00:02:28.520 --> 00:02:34.080
<v Speaker 2>we are running pretty decent sized GP cloud cluster, and

38
00:02:34.120 --> 00:02:36.719
<v Speaker 2>we also through our partners, we bring in some of

39
00:02:36.719 --> 00:02:39.639
<v Speaker 2>the advanced GPS. So before we get that, let me

40
00:02:40.159 --> 00:02:43.960
<v Speaker 2>tell you my journey how it started into AM One

41
00:02:44.039 --> 00:02:47.439
<v Speaker 2>day after my marriage, when everyone was watching my marriage video,

42
00:02:48.080 --> 00:02:52.520
<v Speaker 2>I was in my Mac and reading through the TensorFlow documentation.

43
00:02:52.719 --> 00:02:55.879
<v Speaker 2>That Google announced Sensorflow one day on my problem on

44
00:02:55.919 --> 00:02:58.719
<v Speaker 2>my day of marriage, and I was like thinking that

45
00:02:58.800 --> 00:03:00.599
<v Speaker 2>I had a new Mac back then, and having a

46
00:03:00.639 --> 00:03:03.039
<v Speaker 2>Mac was a big deal back then in India, not

47
00:03:03.199 --> 00:03:06.319
<v Speaker 2>many had it, and I was thinking that, oh, I

48
00:03:06.360 --> 00:03:08.680
<v Speaker 2>should be able to run this the TFM and I

49
00:03:08.680 --> 00:03:12.120
<v Speaker 2>st on my laptop. I tried doing that and the

50
00:03:12.159 --> 00:03:16.120
<v Speaker 2>Mac started crying badly. Was it got very hot. That's

51
00:03:16.159 --> 00:03:19.840
<v Speaker 2>how I got introduced to the concept of GPUs and

52
00:03:20.039 --> 00:03:22.800
<v Speaker 2>me being a great fan of Terminator movie, I wanted

53
00:03:22.840 --> 00:03:26.800
<v Speaker 2>to get into the AI. Tried understanding couple of things,

54
00:03:27.400 --> 00:03:29.639
<v Speaker 2>but dropped it a couple of times, thinking that it's

55
00:03:29.919 --> 00:03:33.599
<v Speaker 2>more for PhD, not my kind of stuff, but did

56
00:03:33.599 --> 00:03:36.680
<v Speaker 2>not give up. It was still curious and then luckily

57
00:03:36.719 --> 00:03:40.520
<v Speaker 2>I came across some of really good content by people

58
00:03:40.599 --> 00:03:43.439
<v Speaker 2>like Jeremy Howard from fast A, where some of the

59
00:03:43.439 --> 00:03:46.439
<v Speaker 2>things started getting clear. I was able to build things,

60
00:03:46.520 --> 00:03:48.879
<v Speaker 2>and luckily on the way, I got introduced to Cagle,

61
00:03:49.240 --> 00:03:52.080
<v Speaker 2>where I started competing in Cagle competitions. Was able to

62
00:03:52.120 --> 00:03:54.360
<v Speaker 2>get into the top of the leaderboard top one percent

63
00:03:54.400 --> 00:03:58.680
<v Speaker 2>in the world, and luckily got some consulting assignments for

64
00:03:58.840 --> 00:04:03.639
<v Speaker 2>companies Fortune and mid sized companies where help companies build

65
00:04:03.639 --> 00:04:06.199
<v Speaker 2>their implementation. It was not called a back then. It

66
00:04:06.280 --> 00:04:09.520
<v Speaker 2>was usually a lot of deep learning, which kind of

67
00:04:09.919 --> 00:04:12.159
<v Speaker 2>morphed into AGE and and all these stuff, but the

68
00:04:12.199 --> 00:04:16.040
<v Speaker 2>basics still remained the same. During this stint, I realized

69
00:04:16.079 --> 00:04:20.759
<v Speaker 2>that the GPU's the GPU compute is provided by the

70
00:04:20.839 --> 00:04:25.000
<v Speaker 2>hyperscalers are very complicated to use and very expensive. Some

71
00:04:25.040 --> 00:04:27.319
<v Speaker 2>of the companies in the US were trying to address this,

72
00:04:27.399 --> 00:04:29.519
<v Speaker 2>but they were still expensive for a lot of people

73
00:04:30.439 --> 00:04:33.399
<v Speaker 2>out there and countries like India and other places. So

74
00:04:33.639 --> 00:04:35.519
<v Speaker 2>I really thought that why not we do it in

75
00:04:35.519 --> 00:04:37.879
<v Speaker 2>India and make it cheaper and make it super simple.

76
00:04:39.040 --> 00:04:43.120
<v Speaker 2>That's how the journey started off. My current company called Javaslabs,

77
00:04:43.160 --> 00:04:47.040
<v Speaker 2>and the company name was inspired by the Javas from

78
00:04:47.199 --> 00:04:52.720
<v Speaker 2>Marvel Movies, which is an AA probably which Sam Adman

79
00:04:52.800 --> 00:04:56.079
<v Speaker 2>is trying to build right the AGI out the world,

80
00:04:56.079 --> 00:04:59.040
<v Speaker 2>which is like super smart, does everything and all. Is

81
00:04:59.079 --> 00:05:02.959
<v Speaker 2>inspired by that. Started this company around twenty nineteen during

82
00:05:03.000 --> 00:05:06.480
<v Speaker 2>the COVID. Before the COVID started, the COVID hit us badly.

83
00:05:06.879 --> 00:05:09.439
<v Speaker 2>We got stuck in place for more than eight months

84
00:05:09.439 --> 00:05:13.560
<v Speaker 2>without able to do anything. But then things started slowly

85
00:05:13.600 --> 00:05:16.040
<v Speaker 2>picking up and now we are doing pretty well.

86
00:05:16.240 --> 00:05:19.600
<v Speaker 1>Awesome, So, mich I mean, based on what you said,

87
00:05:19.639 --> 00:05:21.839
<v Speaker 1>I was going to ask you what is your inspiration?

88
00:05:21.879 --> 00:05:24.240
<v Speaker 1>You kind of touched upon your inspiration. What I gathered

89
00:05:24.360 --> 00:05:28.680
<v Speaker 1>is obviously your dad helped you early on with eunuchs

90
00:05:28.759 --> 00:05:32.839
<v Speaker 1>and whatnot, with your first computer. Then looks like your

91
00:05:32.959 --> 00:05:36.720
<v Speaker 1>marriage video was the cause of you getting inspired for

92
00:05:37.360 --> 00:05:41.399
<v Speaker 1>as well. And then of course terminators. So three people

93
00:05:41.439 --> 00:05:45.279
<v Speaker 1>you need to thank for for your journey. I guess right,

94
00:05:45.360 --> 00:05:46.879
<v Speaker 1>anyone I missed for your journey?

95
00:05:47.439 --> 00:05:50.480
<v Speaker 2>Yeah, I mean I take inspiration from a lot of people. Here,

96
00:05:51.480 --> 00:05:55.720
<v Speaker 2>Jeremy Howard from Fast Andredge carpaty. If it's not for

97
00:05:55.759 --> 00:05:59.920
<v Speaker 2>a Stanford open source curriculum, I have not picked up.

98
00:06:00.360 --> 00:06:02.839
<v Speaker 2>Yeah right, there are a lot of inspirations the way

99
00:06:02.959 --> 00:06:06.759
<v Speaker 2>move on, how to do business, bootstrapped way, how to

100
00:06:06.839 --> 00:06:09.680
<v Speaker 2>think long term. I take inspiration from a lot of people.

101
00:06:09.920 --> 00:06:13.959
<v Speaker 1>Yeah. Yeah, that is amazing because I tell myself this

102
00:06:14.040 --> 00:06:16.600
<v Speaker 1>as well. I mean, you know you you you cannot

103
00:06:16.639 --> 00:06:20.279
<v Speaker 1>shy away from taking inspiration even from a little child.

104
00:06:20.439 --> 00:06:22.240
<v Speaker 1>Or there are times when I look at my dog

105
00:06:22.319 --> 00:06:24.399
<v Speaker 1>do certain things and I'm like, I mean, if if

106
00:06:24.399 --> 00:06:27.079
<v Speaker 1>the dog think that way, there's something in it for

107
00:06:27.160 --> 00:06:28.920
<v Speaker 1>us to learn from that as well, you know, so

108
00:06:29.160 --> 00:06:29.639
<v Speaker 1>I love.

109
00:06:29.519 --> 00:06:32.120
<v Speaker 2>That absolutely right. The dogs can teach us a lot.

110
00:06:34.079 --> 00:06:36.879
<v Speaker 1>Yeah, So let's let's go back a little bit in

111
00:06:36.920 --> 00:06:41.240
<v Speaker 1>your journey vision. I mean, obviously you're having fun doing

112
00:06:41.600 --> 00:06:47.439
<v Speaker 1>Jarvis Labs right now, but looking at your profile, looking

113
00:06:47.519 --> 00:06:49.480
<v Speaker 1>at what you've done, I mean you've done a lot

114
00:06:49.519 --> 00:06:51.560
<v Speaker 1>on the data side of things, right, I mean you know,

115
00:06:51.920 --> 00:06:57.519
<v Speaker 1>whether it's HADOO or analytics or even business intelligence for

116
00:06:57.560 --> 00:07:00.839
<v Speaker 1>that matter. I mean, so you data was kind of

117
00:07:00.879 --> 00:07:05.439
<v Speaker 1>in your ingrained in your blood, or from the beginning

118
00:07:05.519 --> 00:07:08.879
<v Speaker 1>of your career, that's what you were doing. I'm assuming

119
00:07:08.920 --> 00:07:12.399
<v Speaker 1>there was some sort of pivot you wanted or something

120
00:07:12.439 --> 00:07:15.360
<v Speaker 1>that you didn't like or you loved about data, because

121
00:07:15.399 --> 00:07:18.680
<v Speaker 1>I mean, without data, AI is nothing, right. I mean,

122
00:07:18.720 --> 00:07:21.040
<v Speaker 1>you know you put in bad data. I mean you

123
00:07:21.120 --> 00:07:24.480
<v Speaker 1>have bad data. Your AI is as bad as the

124
00:07:24.519 --> 00:07:26.160
<v Speaker 1>data you put in. So can you tell us a

125
00:07:26.199 --> 00:07:29.560
<v Speaker 1>little bit about that journey of where you started you know,

126
00:07:29.639 --> 00:07:32.319
<v Speaker 1>what what in I mean, where you ended up taking

127
00:07:32.360 --> 00:07:36.560
<v Speaker 1>the pivot to this from your data journey, so to speak.

128
00:07:37.959 --> 00:07:40.680
<v Speaker 2>Interesting question. The way I look at it is slightly different.

129
00:07:40.720 --> 00:07:44.079
<v Speaker 2>I'll give you some history towards it. My first job

130
00:07:44.199 --> 00:07:47.120
<v Speaker 2>was in withprow which is a major I service company

131
00:07:47.120 --> 00:07:51.360
<v Speaker 2>in India, and I did not know the difference between

132
00:07:51.360 --> 00:07:55.199
<v Speaker 2>services and products, so I was kind of getting bored

133
00:07:55.240 --> 00:07:57.000
<v Speaker 2>of each of the projectile was in the pro so

134
00:07:57.000 --> 00:07:59.120
<v Speaker 2>I used to fight with my manager and say that no,

135
00:07:59.199 --> 00:08:01.879
<v Speaker 2>I'm not interested to get to something else. So I

136
00:08:01.920 --> 00:08:05.199
<v Speaker 2>started working with business I think it's sap Bo sap

137
00:08:05.199 --> 00:08:09.680
<v Speaker 2>business objects. Then move to Java swings and applet's building

138
00:08:09.759 --> 00:08:12.839
<v Speaker 2>a product for best Buy where probably you guys would

139
00:08:12.839 --> 00:08:15.920
<v Speaker 2>sign it that we did some parts of building that,

140
00:08:16.560 --> 00:08:19.560
<v Speaker 2>so I have not been to best Buy. Then my

141
00:08:19.639 --> 00:08:23.519
<v Speaker 2>last project was MasterCard. I think a lot of things.

142
00:08:23.560 --> 00:08:28.240
<v Speaker 2>So what also happened was until two thousand and twelve

143
00:08:28.319 --> 00:08:32.879
<v Speaker 2>or something, it was primarily proprilatary software sap IBMB software.

144
00:08:32.919 --> 00:08:34.559
<v Speaker 2>So if you want to get into these, you need

145
00:08:34.600 --> 00:08:37.720
<v Speaker 2>to have some account which helps you do it, or

146
00:08:37.720 --> 00:08:39.840
<v Speaker 2>you need to get into an expensive training to get it.

147
00:08:40.320 --> 00:08:42.679
<v Speaker 2>But the moment open so started kicking in in the

148
00:08:42.679 --> 00:08:47.039
<v Speaker 2>form of Android, Spring, Hibernate or how do things started changing.

149
00:08:47.080 --> 00:08:49.240
<v Speaker 2>As long as you have a good Internet, that's why

150
00:08:49.360 --> 00:08:52.919
<v Speaker 2>we got three G in India, So you do one

151
00:08:52.960 --> 00:08:56.919
<v Speaker 2>hundred dollars or Rube reach outs. You get like two

152
00:08:56.960 --> 00:09:00.320
<v Speaker 2>GBO three GB connection. Now you can start download things,

153
00:09:00.360 --> 00:09:02.200
<v Speaker 2>a lot of things. Right. That allowed me to do

154
00:09:02.240 --> 00:09:05.240
<v Speaker 2>get into the open source and when I was working

155
00:09:05.279 --> 00:09:10.000
<v Speaker 2>on business objects, I came across an interesting problem which

156
00:09:10.039 --> 00:09:13.000
<v Speaker 2>was also the root for me jumping into her. So

157
00:09:13.000 --> 00:09:15.120
<v Speaker 2>we had a business objects report which was running for

158
00:09:15.240 --> 00:09:18.759
<v Speaker 2>one of a big client, which started crashing because it

159
00:09:18.840 --> 00:09:21.879
<v Speaker 2>was running for more than an hour, and I got

160
00:09:21.919 --> 00:09:25.679
<v Speaker 2>curious when Google was able to do a search in

161
00:09:25.759 --> 00:09:30.080
<v Speaker 2>like seconds, why does this happen? Why we were not

162
00:09:30.200 --> 00:09:33.480
<v Speaker 2>able to do that. So I namely asked my architect,

163
00:09:33.960 --> 00:09:37.679
<v Speaker 2>who was a very smart lady, why is this happening?

164
00:09:37.720 --> 00:09:40.519
<v Speaker 2>She was also clueless. But then I started digging deeper

165
00:09:40.559 --> 00:09:42.960
<v Speaker 2>into it. That's why I came into this concept called

166
00:09:43.039 --> 00:09:47.399
<v Speaker 2>map reduce. How companies like Google, ads is, distributed computing

167
00:09:47.440 --> 00:09:51.000
<v Speaker 2>and all those That's where my interests started and we

168
00:09:51.120 --> 00:09:54.080
<v Speaker 2>pro luckily got probably one of the earliest projects in

169
00:09:54.080 --> 00:09:56.240
<v Speaker 2>the big data space for MasterCard setting up on one

170
00:09:56.279 --> 00:10:01.080
<v Speaker 2>of their largest big data clusters and cloud pioneer in

171
00:10:01.120 --> 00:10:03.600
<v Speaker 2>that CLOUDA and Horton works and MasterCard was a super

172
00:10:03.600 --> 00:10:06.480
<v Speaker 2>India partner with cloud Era, so we had access to

173
00:10:08.159 --> 00:10:11.240
<v Speaker 2>Clouda engineers I learned. I got an opportunity to learn

174
00:10:11.240 --> 00:10:15.559
<v Speaker 2>a lot from them. That's how my journey into data started.

175
00:10:16.200 --> 00:10:18.360
<v Speaker 2>And I then started moving out. I moved out of

176
00:10:18.360 --> 00:10:21.120
<v Speaker 2>a pro worked in a couple of startups on different

177
00:10:21.159 --> 00:10:24.360
<v Speaker 2>things like reporting, and then I worked on an analytics

178
00:10:24.399 --> 00:10:27.360
<v Speaker 2>company for some time where we were doing things like

179
00:10:27.399 --> 00:10:30.440
<v Speaker 2>Spark and all those things. And Spark was also something

180
00:10:30.480 --> 00:10:33.320
<v Speaker 2>super interesting. The problem with Hadoo was it was amazing,

181
00:10:33.360 --> 00:10:36.519
<v Speaker 2>but the map produce was too complicated, the Java boiler

182
00:10:36.559 --> 00:10:38.799
<v Speaker 2>plate and all this comes. And Spark what it did

183
00:10:38.919 --> 00:10:41.399
<v Speaker 2>is it kind of simplified everything. Whatever you could do

184
00:10:41.559 --> 00:10:44.159
<v Speaker 2>in a much bigger final and mapp produce, you were

185
00:10:44.159 --> 00:10:46.679
<v Speaker 2>able to do in one single line dot map, dot

186
00:10:46.720 --> 00:10:49.600
<v Speaker 2>reduce and do it. Things got simplified over the next

187
00:10:49.600 --> 00:10:53.360
<v Speaker 2>coming years. But that's how I jumped into the Spark

188
00:10:53.759 --> 00:10:57.120
<v Speaker 2>and then Spark had a subset library called machine learning.

189
00:10:57.159 --> 00:11:00.480
<v Speaker 2>That's when I started getting to the came means and

190
00:11:00.639 --> 00:11:03.679
<v Speaker 2>random for us, all those things, and we started implementing

191
00:11:03.759 --> 00:11:06.519
<v Speaker 2>it for some of our customers in the company, and

192
00:11:06.559 --> 00:11:08.759
<v Speaker 2>that's where the tagle thing also started. By we were

193
00:11:08.759 --> 00:11:11.519
<v Speaker 2>able to apply all the open source knowledge that I

194
00:11:11.559 --> 00:11:15.759
<v Speaker 2>was gaining all these years on real problems, challenge and

195
00:11:15.919 --> 00:11:18.320
<v Speaker 2>learn from some of the best in the world. That's

196
00:11:18.320 --> 00:11:22.519
<v Speaker 2>how the pivot happened. And I had to use GPUs

197
00:11:22.559 --> 00:11:25.480
<v Speaker 2>in order to compete in Taglar competitions, which needs a

198
00:11:25.480 --> 00:11:28.360
<v Speaker 2>lot of GPUs. I had one which my friend luckily

199
00:11:29.200 --> 00:11:31.480
<v Speaker 2>got it when he was coming back from you as Titanic.

200
00:11:31.840 --> 00:11:34.960
<v Speaker 2>We were not getting it back then. That's how the

201
00:11:35.000 --> 00:11:35.720
<v Speaker 2>journey started.

202
00:11:36.000 --> 00:11:40.759
<v Speaker 1>Yeah, yeah, So you spoke a little bit about GPUs

203
00:11:40.799 --> 00:11:45.840
<v Speaker 1>and I think it's a nice kind of segue into

204
00:11:45.960 --> 00:11:50.440
<v Speaker 1>what Jarvis Labs is doing right now, right, So maybe

205
00:11:50.559 --> 00:11:53.480
<v Speaker 1>you can spend two minutes on where you've got the

206
00:11:53.519 --> 00:11:57.159
<v Speaker 1>idea or the inspiration for Jarvis Labs, right and what

207
00:11:57.200 --> 00:12:01.120
<v Speaker 1>are you trying to offer to home and you know

208
00:12:01.279 --> 00:12:04.799
<v Speaker 1>who potentially should be looking at Jarvis Labs, So can

209
00:12:04.840 --> 00:12:06.159
<v Speaker 1>you touch on that vision?

210
00:12:08.000 --> 00:12:10.879
<v Speaker 2>Got it? So, as I told you earlier, right when

211
00:12:10.919 --> 00:12:14.399
<v Speaker 2>I was working for one of the large companies, they

212
00:12:14.480 --> 00:12:17.720
<v Speaker 2>were facing one specific problem. They had like huge service

213
00:12:17.759 --> 00:12:20.960
<v Speaker 2>in place with GPS, but the develops team that they

214
00:12:21.000 --> 00:12:24.080
<v Speaker 2>had were not able to address the needs of data scientists.

215
00:12:24.519 --> 00:12:27.200
<v Speaker 2>So what was typically happening was the data scientists wanted

216
00:12:28.080 --> 00:12:31.519
<v Speaker 2>a particular environment. Every time they want installations, they have

217
00:12:31.639 --> 00:12:34.360
<v Speaker 2>to go to develop for permissions. It was the develops

218
00:12:34.399 --> 00:12:36.919
<v Speaker 2>was built for a different kind of a beast, not

219
00:12:37.039 --> 00:12:38.799
<v Speaker 2>for the air world. They did not know how to

220
00:12:38.879 --> 00:12:41.799
<v Speaker 2>handle it, so a lot of compute time was getting wasted.

221
00:12:42.200 --> 00:12:44.320
<v Speaker 2>This was in the bigger companies and in the smaller

222
00:12:44.360 --> 00:12:46.759
<v Speaker 2>companies it was much different, and for individuals it was

223
00:12:46.919 --> 00:12:50.000
<v Speaker 2>much different. It was much harder. Smaller companies cannot even

224
00:12:50.039 --> 00:12:54.759
<v Speaker 2>have this develop steam, so spinning up GPU instances, installing

225
00:12:54.840 --> 00:12:58.360
<v Speaker 2>Nvidia COUDA driver was very hard. So I wanted to

226
00:12:58.720 --> 00:13:01.639
<v Speaker 2>kind of build a company, build a product which kind

227
00:13:01.679 --> 00:13:05.240
<v Speaker 2>of makes this simplified, as I have done that several

228
00:13:05.240 --> 00:13:09.519
<v Speaker 2>times for cable competitions on platforms like AABSGCP. Struggle getting

229
00:13:09.559 --> 00:13:12.960
<v Speaker 2>GPUs in those platforms, even though I was ready to

230
00:13:12.960 --> 00:13:16.720
<v Speaker 2>pay at times, the ex very expensive ones right three hundred,

231
00:13:16.759 --> 00:13:18.440
<v Speaker 2>and if you want to get it at that time,

232
00:13:18.480 --> 00:13:20.759
<v Speaker 2>you need to wait for pretty long. You need approvals,

233
00:13:20.799 --> 00:13:24.039
<v Speaker 2>which you hardly get. There were one or two smaller

234
00:13:24.080 --> 00:13:26.240
<v Speaker 2>players which were like kind of wrappers on these things.

235
00:13:26.600 --> 00:13:30.080
<v Speaker 2>They make it easier, but they were very expensive, not

236
00:13:30.120 --> 00:13:33.960
<v Speaker 2>suitable for the SMB's smaller companies to mid size companies

237
00:13:34.039 --> 00:13:37.200
<v Speaker 2>or for the individuals like us. So that's why the

238
00:13:37.200 --> 00:13:41.559
<v Speaker 2>idea came of Java slabs. Can we put twenty eighty

239
00:13:41.639 --> 00:13:46.600
<v Speaker 2>TI the gaming cards and then build a software layer

240
00:13:46.639 --> 00:13:48.879
<v Speaker 2>on top of it which kind of automates all the

241
00:13:49.039 --> 00:13:52.759
<v Speaker 2>tools that are required for data scientists. So data scientists

242
00:13:52.759 --> 00:13:55.519
<v Speaker 2>can just focus on model building or deployment and not

243
00:13:55.600 --> 00:13:59.320
<v Speaker 2>worry about the software required for the infrastructure. Pad that

244
00:13:59.440 --> 00:14:03.120
<v Speaker 2>Sidea with started the company. After we made some bunch

245
00:14:03.159 --> 00:14:06.639
<v Speaker 2>of initial orders, NBDA made it pick to their licensing

246
00:14:06.720 --> 00:14:09.320
<v Speaker 2>saying that you're not supposed to use this consumer GPUs

247
00:14:09.320 --> 00:14:11.919
<v Speaker 2>in data centers. So we kind of had to take

248
00:14:11.919 --> 00:14:14.080
<v Speaker 2>a pivot from that and say that, okay, let's buy

249
00:14:14.120 --> 00:14:19.039
<v Speaker 2>the commercial GPUs. They're a bit expensive, but the story

250
00:14:19.080 --> 00:14:22.600
<v Speaker 2>remainder the same. So who can use javaslaps If you're

251
00:14:22.720 --> 00:14:26.120
<v Speaker 2>trying to do research, if you're trying to frame your

252
00:14:26.159 --> 00:14:28.960
<v Speaker 2>models on MULTIGPU clusters or even on a single cluster

253
00:14:29.039 --> 00:14:31.799
<v Speaker 2>with agpus or less GPUs, you don't want to get

254
00:14:31.799 --> 00:14:35.240
<v Speaker 2>into monthly, yearly commitments. You want to pages for a minute.

255
00:14:35.519 --> 00:14:38.279
<v Speaker 2>You want to deploy your models very quickly and you

256
00:14:38.360 --> 00:14:40.799
<v Speaker 2>want to have a friendly tech team to support you

257
00:14:40.840 --> 00:14:43.679
<v Speaker 2>in case you'll need something, then probably you should look

258
00:14:43.679 --> 00:14:47.080
<v Speaker 2>at Javislabs. For people that have not really looked into

259
00:14:47.080 --> 00:14:53.000
<v Speaker 2>cloud providers, we are an alternative solution to Google Cloud

260
00:14:53.080 --> 00:14:56.240
<v Speaker 2>or abus. The way I tell some of the vcs

261
00:14:56.240 --> 00:14:59.879
<v Speaker 2>that have met us as GCP is good. They have

262
00:15:00.039 --> 00:15:03.440
<v Speaker 2>a purpose. They are like the ones who manufacture BMW casts,

263
00:15:03.440 --> 00:15:06.360
<v Speaker 2>but unfortunately the entire world cannot buy a BMW or

264
00:15:06.399 --> 00:15:09.759
<v Speaker 2>at Tesla. We need the smaller cast like Marthy Kundai

265
00:15:09.879 --> 00:15:13.000
<v Speaker 2>for different use cases. Also, we offer that so if

266
00:15:13.039 --> 00:15:16.519
<v Speaker 2>you're inn SMB spending huge amounts of money, you can

267
00:15:16.679 --> 00:15:19.360
<v Speaker 2>just switch to a company like Javaslabs for US and

268
00:15:19.559 --> 00:15:23.600
<v Speaker 2>reduce your DevOps cost, increase your efficiency and save a

269
00:15:23.639 --> 00:15:24.320
<v Speaker 2>lot of money.

270
00:15:24.639 --> 00:15:29.200
<v Speaker 1>That's a that's a nice USB. So, given given what

271
00:15:29.279 --> 00:15:31.440
<v Speaker 1>you're offering, I mean, who is using it today?

272
00:15:32.080 --> 00:15:32.320
<v Speaker 2>Can you?

273
00:15:32.360 --> 00:15:35.639
<v Speaker 1>I mean again you don't have to mention names if

274
00:15:35.639 --> 00:15:37.279
<v Speaker 1>you can't, but I mean, at least give us an

275
00:15:37.320 --> 00:15:40.639
<v Speaker 1>idea of who is using and what they're potentially using

276
00:15:40.639 --> 00:15:41.399
<v Speaker 1>it for.

277
00:15:41.639 --> 00:15:44.679
<v Speaker 2>What they're potentially Okay, let's start with who is using it?

278
00:15:45.360 --> 00:15:49.639
<v Speaker 2>Uh So we start Since I was into Cagle. Our

279
00:15:49.679 --> 00:15:53.960
<v Speaker 2>initial focus was people who are doing tagle competitions, right,

280
00:15:54.720 --> 00:15:57.480
<v Speaker 2>we started that. We had a lot of caglers who

281
00:15:57.519 --> 00:16:00.679
<v Speaker 2>actually give us the initial feedback helped us remind product.

282
00:16:01.159 --> 00:16:03.960
<v Speaker 2>But then the problem with Cacalus they are like the competition,

283
00:16:04.039 --> 00:16:06.639
<v Speaker 2>so there's a huge demand peak during the end of

284
00:16:06.679 --> 00:16:10.559
<v Speaker 2>the competition, and we do not have much crowd using

285
00:16:10.600 --> 00:16:13.519
<v Speaker 2>the product in the weekdays or when the competition is

286
00:16:13.559 --> 00:16:17.120
<v Speaker 2>not ending. Right, So we kind of started looking at

287
00:16:17.200 --> 00:16:22.200
<v Speaker 2>more people colleges, labs or people who are offering online courses,

288
00:16:23.320 --> 00:16:26.960
<v Speaker 2>companies who are doing researchers. Some of the top companies

289
00:16:27.480 --> 00:16:30.320
<v Speaker 2>both in the US and India use Java slabs for

290
00:16:30.360 --> 00:16:33.559
<v Speaker 2>their research and for their model deployment purpose. There are

291
00:16:33.639 --> 00:16:36.440
<v Speaker 2>universities from uers where they want to research on building

292
00:16:36.679 --> 00:16:40.279
<v Speaker 2>foundational models or diffusion models or multiple research, right, so

293
00:16:40.279 --> 00:16:42.879
<v Speaker 2>they need clusters of eight hundreds or head hundreds. These

294
00:16:42.879 --> 00:16:44.480
<v Speaker 2>are some of our potential customers.

295
00:16:44.679 --> 00:16:47.960
<v Speaker 1>That's great, and then how long does it take for

296
00:16:48.039 --> 00:16:50.240
<v Speaker 1>them to get started? I mean, is it like an

297
00:16:50.279 --> 00:16:53.279
<v Speaker 1>easy plug and play. Let's say they say they want

298
00:16:53.360 --> 00:16:57.200
<v Speaker 1>to start using it from today to when they can

299
00:16:57.279 --> 00:17:01.799
<v Speaker 1>start using the GPUs and you know, apply a certain

300
00:17:02.559 --> 00:17:06.200
<v Speaker 1>model using Jarvis Labs, I mean, what do you thay,

301
00:17:06.440 --> 00:17:07.680
<v Speaker 1>what do you think is the run made?

302
00:17:07.920 --> 00:17:11.680
<v Speaker 2>Yeah, so we offer two kinds of uh we or

303
00:17:11.720 --> 00:17:14.559
<v Speaker 2>I'll say that we have two kinds of uses. One is,

304
00:17:14.640 --> 00:17:18.119
<v Speaker 2>if you want to use the platform normally, you just

305
00:17:18.279 --> 00:17:20.880
<v Speaker 2>put your credit card and you can start spinning your

306
00:17:20.920 --> 00:17:23.680
<v Speaker 2>first instance and probably less than a minute putting the

307
00:17:23.720 --> 00:17:27.359
<v Speaker 2>card thirty seconds and launching instances like fifteen seconds, and

308
00:17:27.440 --> 00:17:30.039
<v Speaker 2>you can start building your models, downloading your data sets

309
00:17:30.039 --> 00:17:32.720
<v Speaker 2>and all those things. But if I was talking about

310
00:17:32.799 --> 00:17:36.279
<v Speaker 2>other use case where university took a large cluster, that's

311
00:17:36.279 --> 00:17:38.640
<v Speaker 2>why you have like contracts in place. Then you have

312
00:17:38.680 --> 00:17:41.279
<v Speaker 2>the banking transfer and then we block a huge set

313
00:17:41.319 --> 00:17:43.519
<v Speaker 2>of GPUs for them and then they do it. That

314
00:17:43.559 --> 00:17:48.440
<v Speaker 2>could probably take days because of the all the formalities

315
00:17:48.480 --> 00:17:50.599
<v Speaker 2>that we have to go through. For a normal user,

316
00:17:50.640 --> 00:17:52.640
<v Speaker 2>it's usually less than a minute or two. Great.

317
00:17:52.680 --> 00:17:54.960
<v Speaker 1>So the way I understood it, I mean it's almost

318
00:17:55.000 --> 00:17:59.640
<v Speaker 1>like versus a custom build sort of thing, right. I

319
00:17:59.680 --> 00:18:03.400
<v Speaker 1>mean so that so the first one is more swipe

320
00:18:03.400 --> 00:18:05.799
<v Speaker 1>your credit card, create your account, You're ready to go.

321
00:18:06.359 --> 00:18:08.799
<v Speaker 1>Your second one is a little more custom use case

322
00:18:08.880 --> 00:18:14.079
<v Speaker 1>you need a larger you know, processing power, and larger

323
00:18:14.119 --> 00:18:16.599
<v Speaker 1>amounts of data, so they need to work with you

324
00:18:16.720 --> 00:18:19.279
<v Speaker 1>to create a custom project, so to speak, and the

325
00:18:19.480 --> 00:18:23.079
<v Speaker 1>custom set of GPUs assigned for them. So that's a

326
00:18:23.359 --> 00:18:27.200
<v Speaker 1>that's a nice way to probably differentiate. So you said

327
00:18:27.200 --> 00:18:30.640
<v Speaker 1>you started in twenty nineteen, which I mean, where do

328
00:18:30.720 --> 00:18:35.200
<v Speaker 1>you see yourself in the journey and you know, has

329
00:18:35.240 --> 00:18:39.200
<v Speaker 1>it been as per your vision as to where you

330
00:18:39.279 --> 00:18:42.799
<v Speaker 1>are in your journey? I mean, and what are your

331
00:18:42.839 --> 00:18:46.079
<v Speaker 1>plans next? With Jarvis Labs, we have.

332
00:18:46.039 --> 00:18:49.960
<v Speaker 2>Come a long way, for sure, and the Aspace sift

333
00:18:50.039 --> 00:18:53.200
<v Speaker 2>has changed a lot. So our initial focus was towards

334
00:18:54.079 --> 00:18:56.920
<v Speaker 2>a very niche community, very small crowd. The market was

335
00:18:56.960 --> 00:19:00.480
<v Speaker 2>not as big assets today, the competition is not as

336
00:19:00.519 --> 00:19:02.279
<v Speaker 2>fears as it is today, so a lot of things

337
00:19:02.279 --> 00:19:09.400
<v Speaker 2>have changed. People are moving towards contractless GPU compute requirement

338
00:19:09.480 --> 00:19:13.240
<v Speaker 2>or even people are moving towards token based billing right,

339
00:19:13.680 --> 00:19:15.640
<v Speaker 2>so they don't want to get into some of the

340
00:19:15.720 --> 00:19:18.519
<v Speaker 2>use cases. Do not want to get into I want

341
00:19:18.559 --> 00:19:23.200
<v Speaker 2>this XGPO YGP You rather, I want to deploy my

342
00:19:23.359 --> 00:19:25.720
<v Speaker 2>model and you just take care of the entire life

343
00:19:25.720 --> 00:19:28.799
<v Speaker 2>cycle of that instance. Like what I mean is, let's

344
00:19:28.799 --> 00:19:30.880
<v Speaker 2>say you find you in a model which is specific

345
00:19:30.960 --> 00:19:32.920
<v Speaker 2>for your use case. You don't want to keep this

346
00:19:33.000 --> 00:19:36.480
<v Speaker 2>model running twenty four bus seventy sixty l as you

347
00:19:36.599 --> 00:19:39.640
<v Speaker 2>wanted to come up only when there is a request,

348
00:19:39.759 --> 00:19:42.759
<v Speaker 2>just like how Aw'slambda works. Right function as a service.

349
00:19:43.319 --> 00:19:46.440
<v Speaker 2>That's one of the projects which we are actively investing

350
00:19:46.480 --> 00:19:49.200
<v Speaker 2>in it. We are hoping to go live probably in

351
00:19:49.240 --> 00:19:52.839
<v Speaker 2>a month's time, the first version of it, and we're

352
00:19:52.880 --> 00:19:57.400
<v Speaker 2>also seeing that huge research is going on what does

353
00:19:57.519 --> 00:20:01.039
<v Speaker 2>the next thing right. We have evol platform where we

354
00:20:01.160 --> 00:20:03.759
<v Speaker 2>enable companies to do a lot of things, so we

355
00:20:03.799 --> 00:20:06.799
<v Speaker 2>are keeping a closed eye on how the industry is

356
00:20:06.839 --> 00:20:10.559
<v Speaker 2>evolving and we're also making changes according to that to

357
00:20:10.680 --> 00:20:15.920
<v Speaker 2>the platform. Okay, god so I was working My last

358
00:20:15.960 --> 00:20:19.440
<v Speaker 2>company was in Bangalore. I used to work in Chennai.

359
00:20:19.480 --> 00:20:21.839
<v Speaker 2>These two are some of the biggest cities in India,

360
00:20:21.880 --> 00:20:25.440
<v Speaker 2>which probably most of you audiences have heard of. The

361
00:20:25.519 --> 00:20:29.279
<v Speaker 2>challenge with Bangalore is it's a very happening place and

362
00:20:29.319 --> 00:20:31.839
<v Speaker 2>it's also very crowded place. You want to move from

363
00:20:31.920 --> 00:20:35.640
<v Speaker 2>your house to your office and to and fro, it

364
00:20:35.799 --> 00:20:38.000
<v Speaker 2>can take anywhere between two to three hours, even though

365
00:20:38.000 --> 00:20:40.920
<v Speaker 2>the distance is very short. That's one reason I did

366
00:20:40.920 --> 00:20:43.599
<v Speaker 2>not want to continue in Bangalore. The second thing is

367
00:20:44.279 --> 00:20:47.000
<v Speaker 2>I have been a huge fan of Shi the Remainbow

368
00:20:47.799 --> 00:20:50.599
<v Speaker 2>where he has incorporated his one of his offices in

369
00:20:50.640 --> 00:20:54.319
<v Speaker 2>a place called ting Ghasi, which is much smaller than Kayamato.

370
00:20:54.440 --> 00:20:57.880
<v Speaker 2>And the idea of that is a lot of product

371
00:20:57.960 --> 00:21:02.279
<v Speaker 2>companies in the automobile sector, or the companies which manufacture

372
00:21:02.319 --> 00:21:06.799
<v Speaker 2>the rice preparation missions, right, they are in remote parts

373
00:21:06.799 --> 00:21:10.680
<v Speaker 2>of Switzerland or remote parts of Europe, different Germany and

374
00:21:10.759 --> 00:21:13.559
<v Speaker 2>all right, so you need a calm place for you

375
00:21:13.640 --> 00:21:16.319
<v Speaker 2>to focus on building the product rather than getting into

376
00:21:16.440 --> 00:21:20.359
<v Speaker 2>the rat race that the city offers you. So that

377
00:21:20.440 --> 00:21:23.000
<v Speaker 2>inspired me to come to Kuiamatur. And we don't actually

378
00:21:23.039 --> 00:21:25.720
<v Speaker 2>stay in Kaimatur. For people who are native to Kayamato,

379
00:21:25.799 --> 00:21:28.680
<v Speaker 2>they say, oh, you are actually outside or your nearby village.

380
00:21:28.720 --> 00:21:31.079
<v Speaker 2>That's how people used to say, oh that village, But

381
00:21:31.200 --> 00:21:34.079
<v Speaker 2>it's technically part of Kaimatur. It's a place called Kuepu

382
00:21:34.799 --> 00:21:37.839
<v Speaker 2>and it used to be called Miniuti. I'm not sure

383
00:21:37.839 --> 00:21:40.200
<v Speaker 2>if it's still called right now. The climate is really good.

384
00:21:40.240 --> 00:21:45.000
<v Speaker 2>It almost looks like Minioti. Kim is a very clean,

385
00:21:45.240 --> 00:21:48.559
<v Speaker 2>calm place. People are very respectful here. There's a famous

386
00:21:48.640 --> 00:21:53.160
<v Speaker 2>Nisha spiritual place by the tech, yoga, and there are

387
00:21:53.279 --> 00:21:56.200
<v Speaker 2>mountains all around. It's pretty close to Western Ghatsan, so

388
00:21:56.240 --> 00:21:59.960
<v Speaker 2>it's a very peaceful place away from all the happening things.

389
00:22:00.319 --> 00:22:03.519
<v Speaker 2>And it's very closer to banglows like six sevenas to drive,

390
00:22:04.039 --> 00:22:06.039
<v Speaker 2>So when I want to meet someone or when I

391
00:22:06.039 --> 00:22:08.279
<v Speaker 2>want to go for a meetup, I just drive, have

392
00:22:08.400 --> 00:22:10.799
<v Speaker 2>a break, I enjoy the bangalow's nightlife and come back

393
00:22:10.839 --> 00:22:14.519
<v Speaker 2>to And staying in Kimator helps us to focus on

394
00:22:14.559 --> 00:22:19.839
<v Speaker 2>the work and forget about everything else. And the cost

395
00:22:19.920 --> 00:22:21.440
<v Speaker 2>is also way less than in Bangalore.

396
00:22:21.519 --> 00:22:24.200
<v Speaker 1>Yeah, yeah, I mean, you know, I've been a fan

397
00:22:24.279 --> 00:22:27.799
<v Speaker 1>of Committer just personally speaking. You know, back in the

398
00:22:27.880 --> 00:22:30.279
<v Speaker 1>day when I used to be in consulting, I mean,

399
00:22:30.880 --> 00:22:34.720
<v Speaker 1>my company used to have an offshore facility in Comitor.

400
00:22:35.160 --> 00:22:37.559
<v Speaker 1>I used to do a day trip coming into Committer,

401
00:22:37.720 --> 00:22:39.599
<v Speaker 1>but leave in the evening to go to Bangalore for

402
00:22:39.720 --> 00:22:42.720
<v Speaker 1>another visit. I always wondered, I mean, when would I

403
00:22:42.799 --> 00:22:46.119
<v Speaker 1>stay there for a night to spend a little more time.

404
00:22:46.160 --> 00:22:48.720
<v Speaker 1>This time, when I came to meet you, I was

405
00:22:48.799 --> 00:22:51.680
<v Speaker 1>lucky enough to go around for a little bit cour

406
00:22:51.720 --> 00:22:55.519
<v Speaker 1>to see my friend who got me there. So next time,

407
00:22:55.880 --> 00:22:59.400
<v Speaker 1>even more even more time in Coimitto and the suburbs

408
00:22:59.400 --> 00:23:02.759
<v Speaker 1>to learn about more. So I thoroughly enjoyed my trip.

409
00:23:03.240 --> 00:23:06.039
<v Speaker 1>And I mean I hear all the reasons from you

410
00:23:06.160 --> 00:23:12.440
<v Speaker 1>on why you know, especially building something brand new and

411
00:23:12.480 --> 00:23:15.000
<v Speaker 1>a product, sort of a mindset and things like that.

412
00:23:15.119 --> 00:23:17.599
<v Speaker 1>I mean, you need a lot more peace of mind

413
00:23:17.640 --> 00:23:22.279
<v Speaker 1>and quiet that helps you do more with the less

414
00:23:22.319 --> 00:23:25.480
<v Speaker 1>with less time than just worry about Oh, let me

415
00:23:25.519 --> 00:23:27.400
<v Speaker 1>spend two hours to get to work, let me spend

416
00:23:27.440 --> 00:23:30.799
<v Speaker 1>two hours coming back. And that stress itself is is big.

417
00:23:32.039 --> 00:23:36.160
<v Speaker 1>So given that I'm assuming, given you're living in Bangal,

418
00:23:36.279 --> 00:23:38.839
<v Speaker 1>I mean in Comito, you get a good night's sleep,

419
00:23:39.039 --> 00:23:42.680
<v Speaker 1>but I'm sure Jarvis Labs keeps you up at night

420
00:23:42.720 --> 00:23:46.359
<v Speaker 1>as well. On certain things. You know, your roadmap and

421
00:23:46.400 --> 00:23:48.240
<v Speaker 1>what to do next. I mean one of the things

422
00:23:48.240 --> 00:23:52.279
<v Speaker 1>that you worry about our challenges that you see lying

423
00:23:52.279 --> 00:23:52.920
<v Speaker 1>ahead vision.

424
00:23:56.319 --> 00:24:01.000
<v Speaker 2>That's a very interesting question, something very difficult founder to answer.

425
00:24:01.119 --> 00:24:05.200
<v Speaker 2>Probably the founder gets worried about it the worst. Then

426
00:24:05.279 --> 00:24:08.480
<v Speaker 2>we started Jervis Labs, there was little interest in our product,

427
00:24:09.119 --> 00:24:11.799
<v Speaker 2>not many people who are interested in it, so we

428
00:24:11.799 --> 00:24:16.359
<v Speaker 2>were like the dominant player and we didn't have much competition.

429
00:24:16.400 --> 00:24:20.240
<v Speaker 2>People were happy using us. Then a lot of interest

430
00:24:20.359 --> 00:24:22.880
<v Speaker 2>in the VC world came in during the COVID. You

431
00:24:22.920 --> 00:24:26.440
<v Speaker 2>would have seen the zero interest rate heavy VC funding wented,

432
00:24:27.000 --> 00:24:31.000
<v Speaker 2>which also went in subsidizing the services we offer. Some

433
00:24:31.079 --> 00:24:34.680
<v Speaker 2>of our competitors have subsidized the costs a lot, So

434
00:24:34.720 --> 00:24:38.359
<v Speaker 2>that's one challenge that we currently face. The second thing

435
00:24:38.440 --> 00:24:42.480
<v Speaker 2>is we're also seeing the industry getting more mature. The

436
00:24:42.559 --> 00:24:46.519
<v Speaker 2>efficiency is increasing. We also need the There are too

437
00:24:46.559 --> 00:24:50.680
<v Speaker 2>many players today than what were the two years three

438
00:24:50.759 --> 00:24:53.559
<v Speaker 2>years back, and more are coming into it. That's typically

439
00:24:53.559 --> 00:24:57.759
<v Speaker 2>one challenge. Of course, we are kind of addressing that

440
00:24:57.799 --> 00:25:01.119
<v Speaker 2>by not looking them as competitors rather than we are

441
00:25:01.119 --> 00:25:03.880
<v Speaker 2>looking them at as partners. We're partnered with some of

442
00:25:03.920 --> 00:25:07.480
<v Speaker 2>them for supplying us with the infrastructure. For example, we

443
00:25:07.559 --> 00:25:10.279
<v Speaker 2>don't have a base in Europe, but now through one

444
00:25:10.279 --> 00:25:13.119
<v Speaker 2>of the partnerships with one of the largest data centers

445
00:25:13.240 --> 00:25:15.759
<v Speaker 2>over there, we are able to offer heat shendreds in

446
00:25:15.799 --> 00:25:18.519
<v Speaker 2>the Europe region. We're also talking to some of the

447
00:25:18.559 --> 00:25:21.160
<v Speaker 2>companies in the US, so sooner, in the next one

448
00:25:21.240 --> 00:25:24.240
<v Speaker 2>or two months, you'll be able to launch GPUs in

449
00:25:24.279 --> 00:25:27.400
<v Speaker 2>a closer to US region also, so we're looking we're

450
00:25:27.519 --> 00:25:30.480
<v Speaker 2>trying to take that as an advantage. We're also reduced

451
00:25:30.519 --> 00:25:34.519
<v Speaker 2>investing in GPUs because we feel there's abundance of GPS

452
00:25:34.519 --> 00:25:37.880
<v Speaker 2>available suddenly. But thanks to all the VC money came in,

453
00:25:38.039 --> 00:25:39.720
<v Speaker 2>not just we see real estate money and all this

454
00:25:39.839 --> 00:25:43.160
<v Speaker 2>came in. In terms of technology, the technology has been

455
00:25:43.200 --> 00:25:47.480
<v Speaker 2>evolving right now. It's not the small set of people

456
00:25:47.519 --> 00:25:52.960
<v Speaker 2>who want to use AI, doctors, artists, lawyers, advocates. In

457
00:25:53.000 --> 00:25:56.519
<v Speaker 2>the evening, I was talking to elite lawyers who are

458
00:25:56.559 --> 00:25:58.960
<v Speaker 2>trying to bring an AI to ads. Some of their

459
00:25:59.079 --> 00:26:03.279
<v Speaker 2>use cases dispute resolutions, automatic can also we are supporting them.

460
00:26:03.559 --> 00:26:06.759
<v Speaker 2>So the technology has been changing. There are some misunderstandings

461
00:26:06.759 --> 00:26:08.880
<v Speaker 2>of what the technology is capable of, so we are

462
00:26:08.920 --> 00:26:13.279
<v Speaker 2>addressing those challenges. So it's an interesting time. A huge

463
00:26:13.400 --> 00:26:16.680
<v Speaker 2>competition has increased. I think over the next few years

464
00:26:16.680 --> 00:26:21.759
<v Speaker 2>it will be consolidated again thinking long term, keep lean

465
00:26:21.839 --> 00:26:25.319
<v Speaker 2>and survive longer to actually have a better outcome.

466
00:26:25.680 --> 00:26:28.799
<v Speaker 1>You mentioned we see is a few times. I mean

467
00:26:28.799 --> 00:26:32.119
<v Speaker 1>in terms of where you are in the stage of

468
00:26:32.160 --> 00:26:36.359
<v Speaker 1>the company, right, I mean, can you fare a little

469
00:26:36.359 --> 00:26:38.640
<v Speaker 1>bit about you know, I mean, are you you know,

470
00:26:38.759 --> 00:26:42.359
<v Speaker 1>just in seed stage, waiting a little bit before you

471
00:26:42.440 --> 00:26:44.920
<v Speaker 1>go for Series A or Series B. Or what are

472
00:26:44.920 --> 00:26:47.599
<v Speaker 1>your plans in terms of, you know, even getting investors

473
00:26:47.839 --> 00:26:50.200
<v Speaker 1>into the company. I mean you might say you don't

474
00:26:50.240 --> 00:26:53.640
<v Speaker 1>want investors, but I mean just for anyone listening and

475
00:26:53.720 --> 00:26:55.720
<v Speaker 1>interested in Jarvis.

476
00:26:55.440 --> 00:26:58.279
<v Speaker 2>Labs, I mean, what is your message to them? So

477
00:26:58.319 --> 00:27:00.400
<v Speaker 2>we are a bootstrap company. We have not is an

478
00:27:00.440 --> 00:27:04.559
<v Speaker 2>external funding. In the initial few years, we had opportunities

479
00:27:04.559 --> 00:27:06.880
<v Speaker 2>to talk to vcs, but they did not turn postive.

480
00:27:07.440 --> 00:27:10.279
<v Speaker 2>But right now we make enough, we don't need to

481
00:27:10.359 --> 00:27:13.480
<v Speaker 2>raise money and we want to remind that way. I

482
00:27:13.519 --> 00:27:16.720
<v Speaker 2>also believe that remaining bootstep offers a lot of freedom

483
00:27:17.359 --> 00:27:20.799
<v Speaker 2>and I don't want to run behind millions or billions

484
00:27:20.839 --> 00:27:23.000
<v Speaker 2>of dollars. As long as we make enough and as

485
00:27:23.039 --> 00:27:25.279
<v Speaker 2>long as we grow organically, we are happy with it.

486
00:27:25.680 --> 00:27:29.440
<v Speaker 2>One of the advantages of staying in a place which

487
00:27:29.440 --> 00:27:32.359
<v Speaker 2>is far away from the Silicon Valley of India is

488
00:27:32.400 --> 00:27:34.200
<v Speaker 2>also that you don't need to be part of the address.

489
00:27:34.480 --> 00:27:37.799
<v Speaker 2>Right we love building products. We will continue doing it,

490
00:27:37.839 --> 00:27:39.799
<v Speaker 2>but we don't want to run behind the money. We

491
00:27:39.839 --> 00:27:43.240
<v Speaker 2>will make money, but we don't want to run behind it.

492
00:27:43.720 --> 00:27:46.680
<v Speaker 2>So we don't want to go behind we see money

493
00:27:46.720 --> 00:27:54.640
<v Speaker 2>because that it's like taking artificial way of scaling up.

494
00:27:55.319 --> 00:27:57.880
<v Speaker 2>We want to see how we can take inspiration from

495
00:27:57.880 --> 00:28:03.160
<v Speaker 2>companies like Zoo zero from on how we can grow organically,

496
00:28:03.279 --> 00:28:07.599
<v Speaker 2>maintain a healthy lifestyle and build a long lasting company

497
00:28:07.680 --> 00:28:09.880
<v Speaker 2>then a company that is built to be sold to

498
00:28:09.880 --> 00:28:14.880
<v Speaker 2>someone else. So we are not raising funds at least

499
00:28:15.079 --> 00:28:16.799
<v Speaker 2>as much as I can see in the future.

500
00:28:16.920 --> 00:28:19.799
<v Speaker 1>That is super inspiring vision. I mean I wish every

501
00:28:19.880 --> 00:28:23.160
<v Speaker 1>founder can say that, right, But I mean obviously you

502
00:28:23.200 --> 00:28:25.200
<v Speaker 1>see you see you said rat race as well a

503
00:28:25.240 --> 00:28:28.400
<v Speaker 1>few times. It's always a question of comparing one to

504
00:28:28.440 --> 00:28:31.319
<v Speaker 1>the other. You know, what's my evaluation? You know all

505
00:28:31.400 --> 00:28:34.880
<v Speaker 1>that sort of thing. But you you offer a very

506
00:28:34.920 --> 00:28:40.160
<v Speaker 1>refreshing perspective, and I love the references you always make

507
00:28:40.240 --> 00:28:41.200
<v Speaker 1>to Soho and others.

508
00:28:41.240 --> 00:28:42.960
<v Speaker 2>I mean, I know a bunch of guys from.

509
00:28:44.200 --> 00:28:49.319
<v Speaker 1>As well and understand their culture, how they operate. I mean,

510
00:28:49.319 --> 00:28:52.599
<v Speaker 1>if anyone listening here has has not heard about Soho,

511
00:28:52.839 --> 00:28:56.359
<v Speaker 1>I mean that's a super interesting case study and a

512
00:28:56.400 --> 00:29:00.559
<v Speaker 1>big inspiration someone who comes from our state, Vish News

513
00:29:00.559 --> 00:29:04.119
<v Speaker 1>and mine which is Tamil Nadu. So go look up

514
00:29:04.640 --> 00:29:08.240
<v Speaker 1>Zoho as well. So probably the last question for you

515
00:29:08.880 --> 00:29:12.319
<v Speaker 1>is you you worked in a big company, Like you said,

516
00:29:12.440 --> 00:29:16.160
<v Speaker 1>you know you worked in smaller companies as well. Now

517
00:29:16.160 --> 00:29:18.880
<v Speaker 1>you're a founder, you have a team, you have mouths

518
00:29:18.920 --> 00:29:22.799
<v Speaker 1>to feed and build, you know, paychecks to give and

519
00:29:22.839 --> 00:29:29.319
<v Speaker 1>all that. So and in today's generation, you know, probably

520
00:29:29.359 --> 00:29:32.519
<v Speaker 1>different from twenty years ago, where people wanted a paycheck

521
00:29:32.599 --> 00:29:34.960
<v Speaker 1>to take home, they wanted stability. I mean, there's a

522
00:29:35.039 --> 00:29:39.079
<v Speaker 1>lot more kids and college students trying to be founders,

523
00:29:39.160 --> 00:29:43.240
<v Speaker 1>you know, trying to be start their own company, bring

524
00:29:43.279 --> 00:29:46.880
<v Speaker 1>their dream idea to life, and things like that. So

525
00:29:47.519 --> 00:29:50.119
<v Speaker 1>having been through this, I mean, what what what are

526
00:29:50.200 --> 00:29:52.960
<v Speaker 1>your learnings or your advice you would give to somebody

527
00:29:53.000 --> 00:29:55.799
<v Speaker 1>who's in the third year of college or a fourth

528
00:29:55.920 --> 00:29:58.160
<v Speaker 1>year of college trying to come out and be a founder.

529
00:29:58.799 --> 00:30:02.440
<v Speaker 1>What are your nuggets of advice or three nuggets of advice?

530
00:30:04.279 --> 00:30:06.920
<v Speaker 2>I would say, don't wait till you finish your college,

531
00:30:07.160 --> 00:30:10.880
<v Speaker 2>right because when you're in college, you're lucky enough for

532
00:30:11.000 --> 00:30:13.759
<v Speaker 2>your father to be for your mother to be funding

533
00:30:13.799 --> 00:30:16.359
<v Speaker 2>you your life. So you don't need to maintain your lifestyle,

534
00:30:17.000 --> 00:30:19.519
<v Speaker 2>and you don't need to go and convince someone to

535
00:30:19.599 --> 00:30:22.119
<v Speaker 2>work for you or work with you. I prefer work

536
00:30:22.160 --> 00:30:24.519
<v Speaker 2>with you. You can convince your friends to work with you.

537
00:30:24.599 --> 00:30:27.319
<v Speaker 2>So pick up an idea and a lot of things

538
00:30:27.319 --> 00:30:29.759
<v Speaker 2>have changed in the technology space. There is a concept

539
00:30:29.799 --> 00:30:33.240
<v Speaker 2>called ind hackers. While you can build product and with

540
00:30:33.440 --> 00:30:38.400
<v Speaker 2>tools like GPT cloud building products have become a lot

541
00:30:38.519 --> 00:30:40.839
<v Speaker 2>easier if you put enough amount of time and if

542
00:30:40.880 --> 00:30:44.319
<v Speaker 2>you kind of read so you can actually build product

543
00:30:44.400 --> 00:30:47.440
<v Speaker 2>before you finish your finalier. And if you're lucky or

544
00:30:47.480 --> 00:30:50.400
<v Speaker 2>if you're planning enough, you'll be making money also in

545
00:30:50.440 --> 00:30:52.839
<v Speaker 2>the way. So by the time you come out, either

546
00:30:52.920 --> 00:30:55.599
<v Speaker 2>you have the experience building the product or lucky. The

547
00:30:55.640 --> 00:30:58.440
<v Speaker 2>best case scenario is you have built a product and

548
00:30:58.480 --> 00:31:01.119
<v Speaker 2>you also made money, right, so then you'll have a

549
00:31:01.160 --> 00:31:03.880
<v Speaker 2>better way to see whether you want to continue with

550
00:31:03.920 --> 00:31:06.400
<v Speaker 2>that journey or you will know if it's not your

551
00:31:06.400 --> 00:31:10.720
<v Speaker 2>cup of team. Being a founder is very hard. Before

552
00:31:10.759 --> 00:31:13.279
<v Speaker 2>I started Javas Labs, I have heard a lot of

553
00:31:13.279 --> 00:31:16.519
<v Speaker 2>talks oft from cred who also is a founder of

554
00:31:16.720 --> 00:31:20.079
<v Speaker 2>I think free Charge, he sold that or Snapdean or

555
00:31:20.079 --> 00:31:23.759
<v Speaker 2>something like that. He says founder's journey is a lonely journey.

556
00:31:23.880 --> 00:31:26.079
<v Speaker 2>I used to think, oh, that's maybe it's not me,

557
00:31:26.200 --> 00:31:29.480
<v Speaker 2>maybe it's a still But four years into it, I

558
00:31:29.519 --> 00:31:33.160
<v Speaker 2>can say that's true, right, because you have a lot

559
00:31:33.160 --> 00:31:35.319
<v Speaker 2>of things going on in your head which she may

560
00:31:35.359 --> 00:31:36.920
<v Speaker 2>not be able to share with everyone, or you may

561
00:31:36.960 --> 00:31:38.920
<v Speaker 2>not even have the time to share with people, right,

562
00:31:39.720 --> 00:31:42.160
<v Speaker 2>So it's good. It's good to try it in your

563
00:31:42.160 --> 00:31:44.519
<v Speaker 2>college days and see if you want to do it.

564
00:31:45.720 --> 00:31:47.680
<v Speaker 2>For people who come out of college, I would say,

565
00:31:48.200 --> 00:31:51.359
<v Speaker 2>try to get into a particular space which is usually upcoming,

566
00:31:51.599 --> 00:31:54.240
<v Speaker 2>pick up a domain knowledge, and see how you can

567
00:31:54.279 --> 00:31:57.720
<v Speaker 2>solve a problem differently. Watch a lot of videos of

568
00:31:57.839 --> 00:32:00.200
<v Speaker 2>y combinators. There are a lot of videos of See

569
00:32:00.240 --> 00:32:03.160
<v Speaker 2>the Way boves Zero. The learn from these people so

570
00:32:03.200 --> 00:32:06.640
<v Speaker 2>that you will form a kind of insights on what

571
00:32:06.759 --> 00:32:09.240
<v Speaker 2>you want to build, and probably that's a good way

572
00:32:09.240 --> 00:32:13.000
<v Speaker 2>to start. Don't be dependent on VC money to start.

573
00:32:13.519 --> 00:32:15.640
<v Speaker 2>That may or may not happen even if you want to.

574
00:32:16.279 --> 00:32:18.799
<v Speaker 2>Just a small percentage of companies raised V see funding.

575
00:32:19.119 --> 00:32:22.839
<v Speaker 2>A lot of college people I have spoken to, they

576
00:32:22.839 --> 00:32:26.519
<v Speaker 2>expect that I will get funds to build this product

577
00:32:26.559 --> 00:32:30.680
<v Speaker 2>through VC. Usually we see companies don't fund for your

578
00:32:30.720 --> 00:32:35.039
<v Speaker 2>ideas unless until you have like successful exists or you

579
00:32:35.160 --> 00:32:37.559
<v Speaker 2>have a very different background.

580
00:32:37.680 --> 00:32:41.440
<v Speaker 1>Those are great pieces of advisation. I mean, I wish

581
00:32:41.640 --> 00:32:44.000
<v Speaker 1>I'd spoken to someone like you when I was in college.

582
00:32:44.440 --> 00:32:46.640
<v Speaker 1>You know, my first first task when I was in

583
00:32:46.680 --> 00:32:49.200
<v Speaker 1>college was to find a job, settle down or go

584
00:32:49.400 --> 00:32:53.240
<v Speaker 1>to my ms. But times of job definitely change, right,

585
00:32:53.359 --> 00:32:57.839
<v Speaker 1>So your journey is super inspiring, not just for the

586
00:32:57.880 --> 00:33:00.839
<v Speaker 1>next generation but even for people like me. I thank

587
00:33:00.839 --> 00:33:04.759
<v Speaker 1>who've been working for plenty of years. And if you

588
00:33:04.799 --> 00:33:07.359
<v Speaker 1>have an idea, if you have an inspiration, just go

589
00:33:07.519 --> 00:33:10.480
<v Speaker 1>do it. Is you know, is what I learned from you.

590
00:33:11.359 --> 00:33:14.000
<v Speaker 1>So maybe one last question and we can wrap up

591
00:33:14.039 --> 00:33:16.599
<v Speaker 1>this question. I mean, because you're an AI there's a

592
00:33:17.680 --> 00:33:22.839
<v Speaker 1>lot of talk these days about whether AI is just

593
00:33:22.920 --> 00:33:25.599
<v Speaker 1>another this is just another bubble waiting to burst, or

594
00:33:25.640 --> 00:33:28.359
<v Speaker 1>this is for real this time because of all the

595
00:33:28.400 --> 00:33:31.839
<v Speaker 1>you know, money that is being pumped in. There's other

596
00:33:31.920 --> 00:33:36.440
<v Speaker 1>stories about oh there's AI is way is waiting for

597
00:33:36.519 --> 00:33:38.279
<v Speaker 1>a problem to solve?

598
00:33:38.599 --> 00:33:38.799
<v Speaker 2>Right?

599
00:33:38.839 --> 00:33:42.880
<v Speaker 1>I mean, is technology looking for a problem? Having been

600
00:33:42.920 --> 00:33:46.240
<v Speaker 1>in this space much longer than you know, at least

601
00:33:46.240 --> 00:33:51.200
<v Speaker 1>me and many others, and you know you've put in

602
00:33:51.240 --> 00:33:54.759
<v Speaker 1>your days and nights into AI building this company, I mean,

603
00:33:54.799 --> 00:33:58.160
<v Speaker 1>what is your general take on the state of the

604
00:33:58.200 --> 00:34:01.039
<v Speaker 1>industry right now and where do you think it's going

605
00:34:01.119 --> 00:34:01.279
<v Speaker 1>to go?

606
00:34:01.599 --> 00:34:05.640
<v Speaker 2>So AIR is definitely not a bubble, right whether all

607
00:34:05.680 --> 00:34:08.320
<v Speaker 2>the companies all the VC funding will make a profit.

608
00:34:09.000 --> 00:34:14.800
<v Speaker 2>Maybe not. But let's say for hypothetical scenario where all

609
00:34:14.880 --> 00:34:17.800
<v Speaker 2>the funding for AI and all the research work for

610
00:34:17.880 --> 00:34:23.400
<v Speaker 2>a stops right this minute. Right even then, instead of

611
00:34:23.599 --> 00:34:26.639
<v Speaker 2>have needing one hundred people team, you can build a

612
00:34:26.639 --> 00:34:29.679
<v Speaker 2>lot of products with a small three or ten people

613
00:34:29.760 --> 00:34:36.760
<v Speaker 2>company can still do the same. It revolutionizes education. For example,

614
00:34:36.960 --> 00:34:40.719
<v Speaker 2>in Indian space right you have like for thirty or

615
00:34:40.760 --> 00:34:45.039
<v Speaker 2>sixty students one teacher. So with AI that can completely

616
00:34:45.119 --> 00:34:47.519
<v Speaker 2>change if we don't need much advancements in here. Of

617
00:34:47.559 --> 00:34:50.719
<v Speaker 2>course there's a cost involved in it, but assuming that

618
00:34:50.800 --> 00:34:54.639
<v Speaker 2>the cost, engineering is something easier, easier to solve than

619
00:34:55.559 --> 00:34:58.400
<v Speaker 2>going towards the AGI. So I don't think it's a bubble.

620
00:34:58.480 --> 00:35:01.880
<v Speaker 2>A lot can be done with We have just scratched

621
00:35:01.880 --> 00:35:05.320
<v Speaker 2>the surface. That's how I see it. It's not searching

622
00:35:05.360 --> 00:35:07.480
<v Speaker 2>for the problem. It's we have to figure out how

623
00:35:07.519 --> 00:35:10.320
<v Speaker 2>to solve the problems that we have, rather than solving

624
00:35:10.320 --> 00:35:14.760
<v Speaker 2>the fanciest problem. If you're trying to see innovative ways

625
00:35:14.960 --> 00:35:19.800
<v Speaker 2>of how to use it, probably skies. Will it be agier,

626
00:35:19.960 --> 00:35:22.360
<v Speaker 2>that's a completely different thing, which I don't agree and

627
00:35:22.400 --> 00:35:23.559
<v Speaker 2>I don't want that to be true.

628
00:35:23.599 --> 00:35:27.840
<v Speaker 1>Okay, awesome. That's a great insight and a nice perspective

629
00:35:27.840 --> 00:35:30.719
<v Speaker 1>from someone who is breathing this day in and day out.

630
00:35:30.800 --> 00:35:33.760
<v Speaker 1>So thanks so much for that ution. So let's wrap

631
00:35:33.800 --> 00:35:37.159
<v Speaker 1>it up here. I sincerely appreciate you taking the time

632
00:35:37.320 --> 00:35:40.480
<v Speaker 1>in the evening time to jump on this and share

633
00:35:40.519 --> 00:35:45.400
<v Speaker 1>your insights about Jarvis Labs. I'm sure we'll do another episode,

634
00:35:45.519 --> 00:35:47.800
<v Speaker 1>say six months down the line, see looking at your

635
00:35:47.800 --> 00:35:51.440
<v Speaker 1>growth and what else you've learned, but wish thanks so

636
00:35:51.519 --> 00:35:54.239
<v Speaker 1>much for joining us. All the best to you and

637
00:35:54.360 --> 00:35:57.800
<v Speaker 1>Jarvis Labs and your team and we'll stay in touch.

638
00:35:58.000 --> 00:36:00.400
<v Speaker 2>Thank you, Anthony, all the best for you too. Two