Meeting Title: Uttam-Kumaran’s-Personal-Meeting-Room Date: 2024-10-04 Meeting participants: Patrick Trainer, Joshuadeveyra, Nicolas Sucari, Ryan Brosas, Ericson Dalusong, Luke Daque, Uttam Kumaran


WEBVTT

1 00:00:20.700 00:00:21.270 Uttam Kumaran: Nice

2 00:00:22.150 00:00:23.839 Uttam Kumaran: Brian. I got ahead of it.

3 00:00:24.800 00:00:25.880 Luke Daque: Oh, cool!

4 00:00:26.740 00:00:27.510 Ryan Brosas: Nice.

5 00:00:27.510 00:00:29.610 Uttam Kumaran: I’m not using it now. But I

6 00:00:30.430 00:00:37.940 Uttam Kumaran: yeah, I feel like I’m in so many calls all day. So I I was like, let me just get a headset. I think I should get a mic like Erickson, but I don’t know.

7 00:00:39.540 00:00:42.989 Luke Daque: Like Erickson’s, like a Podcaster, or something.

8 00:00:43.940 00:00:44.400 Uttam Kumaran: Know

9 00:00:45.090 00:00:45.849 Uttam Kumaran: I am.

10 00:00:45.850 00:00:46.950 Patrick Trainer: Call Mike too.

11 00:00:47.310 00:00:48.016 Luke Daque: Oh, yeah.

12 00:00:48.660 00:00:49.530 Ericson Dalusong: Yeah.

13 00:00:49.890 00:00:51.150 Ericson Dalusong: 11th at night.

14 00:00:52.790 00:00:57.859 Uttam Kumaran: Yeah, I just I. I have air pods within my ear hurts because I wear the air pods for like 10 h like.

15 00:00:58.080 00:00:58.810 Luke Daque: Yeah.

16 00:00:58.810 00:01:03.707 Uttam Kumaran: Every day. I wear them for like hours and hours and hours, and I don’t think that’s good

17 00:01:04.599 00:01:05.250 Uttam Kumaran: for my help.

18 00:01:05.250 00:01:08.359 joshuadeveyra: Representative go get low, but after like 4 or 5 h.

19 00:01:08.530 00:01:10.429 Uttam Kumaran: No, I have the airpods pro.

20 00:01:10.460 00:01:12.489 Uttam Kumaran: It’s pretty good like

21 00:01:12.990 00:01:15.349 Uttam Kumaran: you could say on, I mean, like

22 00:01:15.430 00:01:18.399 Uttam Kumaran: 10 h will go by, and then I’m like

23 00:01:18.867 00:01:20.879 Uttam Kumaran: they don’t need to charge, or anything.

24 00:01:21.140 00:01:25.149 Luke Daque: That’s why I’m using wired, I guess, and no, no need for.

25 00:01:25.420 00:01:25.960 joshuadeveyra: Yeah.

26 00:01:25.960 00:01:27.300 Luke Daque: Charging, stuff.

27 00:01:29.414 00:01:30.370 Luke Daque: Yeah.

28 00:01:30.890 00:01:33.309 Roy Christian Piñon: Have you tried? Open back with Tom.

29 00:01:34.412 00:01:38.479 Uttam Kumaran: No, I haven’t got any open. I have. I just have closed

30 00:01:38.960 00:01:44.240 Uttam Kumaran: earphones. I got the sony earphones, and I used to wear Bose. These are like on ear.

31 00:01:45.780 00:01:46.789 Patrick Trainer: So! Oh!

32 00:01:47.646 00:01:53.199 Uttam Kumaran: Music with this, I’m like talking. And so it just needs to be comfortable.

33 00:01:53.645 00:01:58.060 Uttam Kumaran: The quality. Yeah. So like, if we’re in meetings where I’m just here all day sitting.

34 00:01:58.870 00:02:00.319 Uttam Kumaran: I feel like it’s qualified.

35 00:02:00.320 00:02:04.309 Patrick Trainer: These I feel like I could sleep in. They’re so comfortable. Yeah.

36 00:02:04.310 00:02:04.840 Luke Daque: Yeah.

37 00:02:05.400 00:02:10.159 Luke Daque: I have those as well. Mine’s like, I don’t know what these are like.

38 00:02:10.169 00:02:10.459 Patrick Trainer: Right.

39 00:02:10.460 00:02:11.270 Luke Daque: Or something? Yeah.

40 00:02:11.270 00:02:12.620 Patrick Trainer: Oh, those are Sennheisers.

41 00:02:12.810 00:02:13.510 Luke Daque: Yeah.

42 00:02:13.510 00:02:15.359 Patrick Trainer: Yeah, those are nice. Yeah, these are the.

43 00:02:15.360 00:02:16.529 Luke Daque: Comfortable as well.

44 00:02:16.790 00:02:17.670 Patrick Trainer: Dynamic.

45 00:02:17.750 00:02:18.550 Luke Daque: Yeah.

46 00:02:19.280 00:02:20.260 Uttam Kumaran: Nice.

47 00:02:21.230 00:02:23.339 Roy Christian Piñon: Patrick could talk me to sleep.

48 00:02:24.740 00:02:26.720 Patrick Trainer: Does it like it? Does it sound good.

49 00:02:26.720 00:02:27.430 Roy Christian Piñon: Yeah.

50 00:02:27.810 00:02:28.870 Ericson Dalusong: Like, almost.

51 00:02:28.870 00:02:29.739 Roy Christian Piñon: Read this story.

52 00:02:29.740 00:02:31.180 Uttam Kumaran: I thought you’re saying it’s boring.

53 00:02:36.790 00:03:01.909 Patrick Trainer: I’ve I’ve a few people have noticed it, and like I was like at my last job is when I got this mic and I was asking people on my team, I was like, Oh, does it like? Does it sound good? Does it sound any different? And like they had, they’re like I don’t, I don’t know, like couldn’t tell. And then and then I’m like, I know, it’s going to make a difference like, just tell me it sounds good. And then like, so

54 00:03:01.930 00:03:06.080 Patrick Trainer: you saying, that makes me feel really good. So so thank you.

55 00:03:06.400 00:03:10.279 Roy Christian Piñon: Good do you have a like? Is that xlr, or that’s.

56 00:03:10.770 00:03:13.120 Patrick Trainer: It’s a USB-c

57 00:03:13.680 00:03:19.019 Patrick Trainer: but it’s using. But it’s it’s so. Oh, yeah, this is USB-c.

58 00:03:19.140 00:03:28.370 Patrick Trainer: and then my headphones are connected to a to a Dac that is connected via Firewire.

59 00:03:30.450 00:03:33.749 Roy Christian Piñon: Sounds good. I mean, just see, okay.

60 00:03:33.750 00:03:35.840 Patrick Trainer: Or thumb. Yeah. Firewire. Yeah.

61 00:03:37.170 00:03:38.109 Patrick Trainer: Yeah. But the.

62 00:03:38.110 00:03:42.009 Roy Christian Piñon: How about you, Eric? I I haven’t heard yours much.

63 00:03:43.472 00:03:48.629 Ericson Dalusong: This is audio technica I’m not sure if this sounds good on your end. But

64 00:03:48.830 00:03:53.399 Ericson Dalusong: you know when I it does sound good, but not

65 00:03:53.520 00:03:55.789 Ericson Dalusong: as good as road, so I love.

66 00:03:55.790 00:04:03.709 Uttam Kumaran: Zoom. Zoom also degrades it. A bit like your local recording will be really good on Zoom. You can go into settings and set up professional audio.

67 00:04:03.710 00:04:04.200 Patrick Trainer: Yeah.

68 00:04:04.510 00:04:14.440 Uttam Kumaran: Help a lot. Same with video like you can set up a lot of your video but that’s also why, like, we’re on zoom all day, and that’s why even I sent a message

69 00:04:14.450 00:04:19.690 Uttam Kumaran: which is like, if you got if anyone wants like, I mean, everybody here, I feel like has good headphones.

70 00:04:19.769 00:04:26.250 Uttam Kumaran: So. But if if anyone’s like I just got these, if anyone’s like, I want better earbuds or camera, let me know

71 00:04:26.580 00:04:31.880 Uttam Kumaran: because I don’t know. It makes a difference like I feel like for us now that we’re all Async like chair desk

72 00:04:31.900 00:04:39.180 Uttam Kumaran: monitors like it matters a ton, otherwise it’s so so hard to get work. It’s just like hard to do everything. So

73 00:04:39.530 00:04:45.969 Uttam Kumaran: I I mean, I still do a lot of stuff on my laptop like solo. But then I also now have, like everything here.

74 00:04:46.060 00:04:48.059 Uttam Kumaran: So if anyone needs anything.

75 00:04:48.060 00:05:01.629 Patrick Trainer: I’ve actually gone back to the using just one screen. I felt like like I had. I used 3 screens at like laptop, and then 2 at 1 point, and I would just have so much stuff open I’d be like

76 00:05:01.700 00:05:04.549 Patrick Trainer: it like it was like information overload

77 00:05:05.001 00:05:07.160 Patrick Trainer: but then going back to

78 00:05:07.310 00:05:10.729 Patrick Trainer: like just one. It I don’t know. It seems

79 00:05:12.180 00:05:12.830 Patrick Trainer: more.

80 00:05:12.830 00:05:20.710 Luke Daque: 3 might be too many like one is also like too little for me, so maybe 2 probably is the the sweet spot.

81 00:05:21.510 00:05:22.000 Nicolas Sucari: Yeah, me, too.

82 00:05:22.000 00:05:25.799 Roy Christian Piñon: There’s a there’s a corresponding meme to that, Patrick. I’ll send it.

83 00:05:26.730 00:05:27.490 Patrick Trainer: Okay.

84 00:05:30.820 00:05:37.560 Uttam Kumaran: Nice. Well, I didn’t send an agenda today. We just had a lot of stuff go on this week, so maybe I’ll just give a brief

85 00:05:38.162 00:05:39.640 Uttam Kumaran: recap. I think

86 00:05:39.680 00:05:52.742 Uttam Kumaran: the usual schedule. I’m trying to get into a good flow for these and Mondays. We’ve changed these formats a lot, but I think we’re in a pretty good format. Basically, I’ll give a little bit of update about maybe the business as a whole.

87 00:05:53.540 00:06:16.239 Uttam Kumaran: yeah, I think the probably the most stressful parts of life for me, or like the week before month close, basically because that’s where we’re basically like running out of money, and then money is coming in from clients like this week and next week so for me, that’s like where I just make sure that, like we have to close a bunch of books for

88 00:06:16.713 00:06:20.919 Uttam Kumaran: accounting. We have to send invoices out and things like that. So

89 00:06:21.180 00:06:44.950 Uttam Kumaran: after that’s done, I like. I’m happy again for like 2 weeks. So we’re we’re we’re month is closed. Last month was really good. I think a lot of the stuff that I’ve set out to do since July, which was in July, like early August. The big focus was me was like, build our sales and content engine and I think, you know, we’ve really done an amazing job at that.

90 00:06:45.286 00:07:00.449 Uttam Kumaran: You know, I think on the content side, we have content going on all platforms now. Basically everything except for threads, I think. But I don’t think anybody uses that anyways. So we have. We’ve published a tiktok this week we have Instagram.

91 00:07:00.450 00:07:19.470 Uttam Kumaran: We have Linkedin going. My personal Linkedin is going. My personal Twitter is really growing. We have our newsletter go out additionally on the website side. We have the. We have the homepage. That’s almost done. We have redesigns for the blog

92 00:07:19.470 00:07:37.280 Uttam Kumaran: and for the about us done. We’ve we’ve published 2 blog posts so far, and we’re pretty consistent now in our blog schedule, the last blog post we posted. We had great illustrations as well. So I would say already, we’re basically leading in terms of having great illustrations on like predictable posts.

93 00:07:37.681 00:07:46.690 Uttam Kumaran: So on the content side, we’re doing a great job. We’re starting to learn a lot about outbound marketing. Erickson’s, you know, we’re sending emails from a bunch of accounts

94 00:07:46.700 00:08:03.569 Uttam Kumaran: to targets. We’re iterating and learning. How do we target the right people? And I think I’m I’m hopefully trying to do my best job at giving feedback and then also learning like what our process is. I know it’s a big expense for stuff now. But I ideally, as we kind of

95 00:08:03.730 00:08:15.079 Uttam Kumaran: get into the right cadence and we get leads in one, it’ll pay for itself but 2. We can optimize what we need and what we don’t need. So I’m super excited for that on the

96 00:08:15.500 00:08:34.109 Uttam Kumaran: what else? We talked about marketing on the sales side. So we have 3 clients right now that are active. We have by the cocoa that we’re working with. We have full parts and we have Javi coffee. We are next week gonna start an engagement with real data.

97 00:08:34.406 00:08:55.170 Uttam Kumaran: They’re bringing us on to one of their clients to service. Ryan, pat and Nico, I just sent a bunch of notes in there. So if you can get back to me with that, we want to kind of kick stuff off next week. But basically kind of what we’re doing for them is we’re real, has people that on board onto their product. But we

98 00:08:55.170 00:09:13.420 Uttam Kumaran: are like one of the few agencies that do real development. And so they want to partner with us to help onboard their clients onto their product itself, because they don’t have sales engineering internally. So, Roy, you might be interested is like this is an interesting angle that we wanted to do is where we partner with actual some of the vendors. And

99 00:09:13.420 00:09:34.579 Uttam Kumaran: not only do we potentially get referrals from bringing new clients to them. But sometimes they bring us on because we’re the subject matter experts, and we can work quickly. And so this is a great opportunity, I think, to get closer with real and we just have a preliminary agreement for this one engagement. I think I mean we’re gonna knock out of the park, and then kind of have a couple more. So ideally by

100 00:09:34.600 00:09:37.982 Uttam Kumaran: next week we should be up to 4 clients. We’re also

101 00:09:38.770 00:09:43.419 Uttam Kumaran: probably doing a demo at the end of this month for a 5th client called Hpi

102 00:09:43.819 00:09:54.169 Uttam Kumaran: Hpi is, I think I mentioned them before. But they are the number one commercial real estate company here in Austin. So they own, like

103 00:09:54.460 00:10:12.850 Uttam Kumaran: hundreds, probably like hundreds of thousands, maybe like more than a million square feet here in Austin of building. So they own buildings, and they do property management for buildings. We are putting together a demo that Miguel has been working on about helping them build an AI lease negotiation assistant.

104 00:10:13.270 00:10:36.630 Uttam Kumaran: What that does is as they negotiate leases with their tenants. They have a hub where they can put in those requested changes. Basically, take their past leases and account for what they should, what recommendations they should be doing. For example, when a client wants to waive their security deposit, add an extension, change a clause. The AI is helping them look at their past leases and propose a solution.

105 00:10:36.680 00:10:50.520 Uttam Kumaran: So that should, that’s a massive project. That isn’t gonna be a typical engagement. We’re actually pitching them on like a project based fee and that ideally we will. If we get that project, we’ll probably need to bring on

106 00:10:50.999 00:11:16.350 Uttam Kumaran: another person on the AI side, which we’re currently recruiting, for. I’ll talk a little bit about that. And probably someone dedicated to like managing that product and like product management capacity that can deliver, get feedback and things like that. But we’ll see a lot of that hinges on that product. But the Demos have been really great. We’ve learned a lot in the process. On the AI side. So I’m super super excited.

107 00:11:16.980 00:11:19.910 Uttam Kumaran: on the recruiting side. So we are also

108 00:11:19.940 00:11:21.035 Uttam Kumaran: basically

109 00:11:22.270 00:11:41.420 Uttam Kumaran: in advance of getting sales going. We need to make sure we have the right people. So we’re we’re always in a recruiting mode. But we’re we’re particularly interested in in lining up. You know, talented candidates. For open requirements here all the time. In particular, we’re looking for great

110 00:11:41.420 00:11:58.600 Uttam Kumaran: data analysts and data. Engineers. On the data analyst side and data engineering side. I’ll I’m happy to share the job descriptions with everybody here. If there’s anybody in your world that fits it even slightly. Please send it to us, and we’re happy to have a conversation with them.

111 00:11:58.973 00:12:02.859 Uttam Kumaran: We’re willing to hire wherever we’re just looking for the best people.

112 00:12:03.229 00:12:14.060 Uttam Kumaran: And so we’re we’re interviewing with Nico’s interviewing one candidate next week. I had a great interview with someone that Miguel put me in touch with on the AI side.

113 00:12:15.390 00:12:23.440 Uttam Kumaran: we’re starting to work with some external partners who I’ve been in touch with over the past few months about if they have any candidates.

114 00:12:24.130 00:12:37.230 Uttam Kumaran: so that’s that’s gonna be our growth plan again. We’re never like. I’m not in the mode of ever trying to be put in a position to have to hire on the engineering side. I think. We have other.

115 00:12:37.280 00:12:54.659 Uttam Kumaran: We we have capacity internally, and we have some people who are part time, who can come in and do work. We’re we’re looking for the best people and people that we can build long term relationships with, you know. I’ve said that to everybody here on this call, you know, since we’ve all started working together that this isn’t a place

116 00:12:54.660 00:13:14.420 Uttam Kumaran: where I’m I’m trying to bring up, you know. When I was in a startup I was like I joined, and I was immediately like cool like in 2 years I can leave like that’s not. That’s I mean. That may be the case for folks, but that’s not what my goal is. My goal is a place that you know. It’s a place where you can really grow your career, and there’s a whole host of stuff that you can do here and manage and change

117 00:13:14.750 00:13:26.739 Uttam Kumaran: and so we’re looking for people that want to work with us long term, you know. Not just like one month here, one month there folks that want to grow and so that’s that’s kind of the the stuff we’re doing on the recruiting side.

118 00:13:27.258 00:13:28.722 Uttam Kumaran: In terms of

119 00:13:29.220 00:13:46.121 Uttam Kumaran: AI! Miguel did a did a bunch of stuff this week around rag so we built some stuff where we can use rag on top of documents and Google drive. We’re using some new tools to for vector stores. And we’ve kind of had an interesting

120 00:13:46.520 00:13:54.639 Uttam Kumaran: you know, couple of things that I need to review this today. But some great demos there. I don’t know, Miguel. Do you want to pull up like

121 00:13:55.896 00:14:10.243 Uttam Kumaran: any screenshots you have on that or just one of the screenshots. Maybe you sent me, because not everyone is going to be familiar with anything I just said there. So I just want to give people like a basic overview of like what we’re talking about. And and I’ll see also talk about why we

122 00:14:10.800 00:14:12.820 Uttam Kumaran: why, we’re exploring that right now.

123 00:14:13.270 00:14:17.499 joshuadeveyra: Okay, yeah, sure, sure, let me, actually, I’ll just share screen. I think.

124 00:14:19.602 00:14:21.650 joshuadeveyra: can you guys see my screen? Yeah.

125 00:14:22.440 00:14:29.919 joshuadeveyra: okay, so basically, the reason why we’re exploring the these and not the relevance, is because

126 00:14:30.060 00:14:32.620 joshuadeveyra: relevance is a really terrible

127 00:14:33.940 00:14:40.790 joshuadeveyra: why am I logged out? But whatever I they have a really terrible rag, right? So it’s really bad.

128 00:14:41.310 00:14:47.989 joshuadeveyra: So we explored N. 8 N. With, it’s actually I only have N. 8 in my laptop dressed around my PC,

129 00:14:48.090 00:15:05.359 joshuadeveyra: but basically, what happens is that we can, instead of us having to manually, you know, drop everything into relevance and move back and forth. Now, we can just put the file in. If it’s a meeting notes or whatever a summary. We just put that in a Google drive and it automatically converts that into

130 00:15:06.300 00:15:16.540 joshuadeveyra: a post. Vector dB, something I don’t know. They it told me to write some SQL code. Good good thing, and then it and it and provided it. So I didn’t really

131 00:15:16.630 00:15:26.990 joshuadeveyra: do it. That much. So yeah. So for example, here, I uploaded like, I think this is around 90 or almost a hundred pages of lease agreements

132 00:15:27.040 00:15:29.580 joshuadeveyra: and basically just changes. So

133 00:15:30.320 00:15:35.709 joshuadeveyra: I specifically asked for parking space because this was like on the end of the documents.

134 00:15:36.230 00:15:46.689 joshuadeveyra: and it is pretty accurate. 23 is accurate, 18, and even like, for example, the increase after 3 years. So it is working perfectly.

135 00:15:47.272 00:15:52.269 joshuadeveyra: So, yeah, well, the plan is to have this for each sales hub, and you know

136 00:15:52.300 00:15:53.319 joshuadeveyra: that one.

137 00:15:53.620 00:16:05.489 joshuadeveyra: And then the other thing I guess I have to mention with them is the azure open. AI. I’m still exploring this. I actually broke it earlier. But good thing, you know I was able to revert it back. But yeah.

138 00:16:06.690 00:16:30.259 Uttam Kumaran: Yeah. So one of the things that we’re building is every lead that we work on. Right? So basically, our the marketing funnel is we go from like someone who’s curious to someone who books a meeting as soon as they book a meeting, we basically want to create like a sales hub for them. There are some tools that do this, but they’re very, very expensive. And we’re we’re basically gonna do this for like pennies in the dollar

139 00:16:30.612 00:16:58.399 Uttam Kumaran: but basically, we want to set up like a Google drive that has contracts, documents, meetings, anything about that client that waste allows us to manage like 5 or 10 different leads at any moment, which is one of the difficulties that, like I have is what’s the context of what we talked about, what are the feedback they get? And so for us internally, we’re building sales, hubs for every lead, and then, of course, for every client that we work on so we can continue to put in contacts, repos

140 00:16:58.659 00:17:24.620 Uttam Kumaran: slack channels, everything. And then, whenever you want to ask a question about a client you can ask, and it’ll refer back from our documents right? The way we do this now is we have some clod projects, and we have some fixed prompts, and you have to upload docs in there. You have to get the doc. We wanna take care of all that. So, for example, like, if we had a meeting like here, where we the the end state of this is like, let’s say, I have a meeting booked with with buy to coco.

141 00:17:24.760 00:17:46.920 Uttam Kumaran: We we can actually look at the emails that are in the meeting identify that it is a vitaco related meeting automatically, send the transcription and video to the drive, and then it automatically becomes part of context. And so that’s kind of like what you can think about. This process is basically the we want to facilitate the movement of this information about a client to the right area.

142 00:17:46.920 00:18:02.499 Uttam Kumaran: and for folks on the sales side. You know, and even folks like folks like Ryan on the content side. We want to have all of our brain forge information in one place. So you’re not always so because I’m the basically the main blocker of like, what are our past things we’ve done? Tell me about this client

143 00:18:02.500 00:18:29.499 Uttam Kumaran: all that I want to make accessible to everybody. So you can ask the chat to tell you about the client we worked with, and it can prefer straight from the the documents. So we’re gonna build one for every lead that’s open, every existing client, and one for us. With all of our information, we’ll learn a little bit about how do you pipe slack into this zoom, into this everything. And then we’re also basically through this process, also developing this proof of concept for Hpi.

144 00:18:29.860 00:18:33.769 Uttam Kumaran: so a lot of like amazing updates on the AI side, I think.

145 00:18:34.830 00:18:40.100 Uttam Kumaran: probably in the next month or so we’ll we’ll publish some content on what we learned through this process.

146 00:18:40.450 00:18:47.410 Uttam Kumaran: But we’re using a lot of cutting edge tools, a lot of which I just found, you know, kind of scrolling on Twitter and interacting with people there. So

147 00:18:48.830 00:18:52.109 Uttam Kumaran: yeah, that’s the main stuff on the AI side. Any questions

148 00:18:52.760 00:18:54.459 Uttam Kumaran: on this for Miguel.

149 00:18:56.580 00:19:07.309 Roy Christian Piñon: Were you thinking of more into Google drive, or did you have a different space? Or would it be notion? How do you imagine the sales hub to be.

150 00:19:08.020 00:19:09.845 Uttam Kumaran: Yeah. So for me, the

151 00:19:10.440 00:19:22.270 Uttam Kumaran: what I realize about the AI stuff is like the medium really matters right? Like you. Let’s say we get this all working. But you have to log into some like esoteric system to like upload and do stuff. It’s never gonna get beyond.

152 00:19:22.520 00:19:32.610 Uttam Kumaran: Like Miguel and me, basically, this is something that we want everybody to use. So that’s why I wanted to use Google drive, because everybody here has access to Google drive. I want it to be as easy as

153 00:19:32.780 00:19:39.340 Uttam Kumaran: you. Just upload whatever you want for the context, and then you can go to the chat Bot, and the middle part is taken care of.

154 00:19:39.828 00:19:46.059 Uttam Kumaran: And so Google drive is a great medium for everybody else, for us, for on storage. And then also

155 00:19:46.582 00:19:49.820 Uttam Kumaran: we will be automatically moving the notion

156 00:19:49.850 00:19:51.589 Uttam Kumaran: data into there as well.

157 00:19:51.660 00:19:57.029 Uttam Kumaran: So I don’t like using like unfor. Unfortunately, like, we have now multiple tools with this information.

158 00:19:57.080 00:19:58.130 Uttam Kumaran: It’s just

159 00:19:58.150 00:20:08.039 Uttam Kumaran: operational complexity that’s hard to kind of avoid. But I want to use Google drive as a storage layer, and then we will end up piping all of the information into there.

160 00:20:10.180 00:20:13.609 Uttam Kumaran: You can expect, probably like daily or something like that, basically

161 00:20:14.030 00:20:22.659 Uttam Kumaran: cause. There, you could tell there’s multiple use cases right like, let’s say we have this meeting right now. And you wanna probably like late tomorrow, you want to say, Hey, tell me a little bit about what we talked about

162 00:20:22.790 00:20:29.230 Uttam Kumaran: that you have to go into zoom. Or right now we have some of that working, and it goes into notion. But this is the next step right?

163 00:20:29.290 00:20:43.270 Uttam Kumaran: And so basically, I want to reduce any sort of information. Recall where you have to summarize or figure out what we said, or learn something about the company. All of that we have written somewhere, and I want that to all be available via

164 00:20:43.370 00:20:45.890 Uttam Kumaran: one or more chat bots.

165 00:20:46.650 00:20:52.909 Uttam Kumaran: And once we do this internally right, we’ll find out how this all works, and we’ll be able to market and and sell

166 00:20:56.220 00:20:57.400 Uttam Kumaran: cool, and then.

167 00:20:57.400 00:20:58.310 Luke Daque: Awesome.

168 00:20:58.770 00:20:59.100 Uttam Kumaran: Yeah.

169 00:20:59.100 00:21:04.789 Luke Daque: For for the N. 8 N. What language, what Llm is it using currently.

170 00:21:05.130 00:21:05.860 Luke Daque: is it like.

171 00:21:05.860 00:21:06.300 joshuadeveyra: Oh, okay.

172 00:21:06.300 00:21:07.000 Luke Daque: AI!

173 00:21:07.430 00:21:11.560 joshuadeveyra: Yeah, yeah, it’s just using openai. But we want to migrate to this one.

174 00:21:12.150 00:21:13.290 joshuadeveyra: Free credits.

175 00:21:13.610 00:21:17.270 Uttam Kumaran: Yeah. So I applied, I applied for azure startup. And we got like

176 00:21:17.410 00:21:22.289 Uttam Kumaran: a bunch we got like 25 grand in free credit. So I just basically was like, I don’t want to pay for

177 00:21:22.740 00:21:28.301 Uttam Kumaran: we just have to move the the Api keys over. But yeah, hopefully, it’s as simple as that.

178 00:21:29.860 00:21:34.879 Luke Daque: But are we also like like in the settings? Can we change like the

179 00:21:35.210 00:21:41.860 Luke Daque: what do you call it? The version of the o of open AI, for example, like from 3.5 to 4 0. And stuff.

180 00:21:41.860 00:21:46.969 Uttam Kumaran: Yeah, I think when you, when he creates the open AI service, I think you’ll be able to pick which model you use.

181 00:21:46.970 00:21:48.209 joshuadeveyra: So here you can.

182 00:21:48.210 00:21:48.830 Uttam Kumaran: Yeah.

183 00:21:48.830 00:21:49.590 Luke Daque: Nice.

184 00:21:49.590 00:21:51.849 joshuadeveyra: This one is not very simple.

185 00:21:53.030 00:21:54.050 joshuadeveyra: Yeah.

186 00:21:54.050 00:21:56.517 Uttam Kumaran: It’s gonna be a bit more complicated.

187 00:21:57.900 00:21:58.549 Uttam Kumaran: but it’s.

188 00:21:58.550 00:22:01.779 Luke Daque: That’s cool. Yeah, it’s it’s pretty cool, though.

189 00:22:01.990 00:22:08.120 Luke Daque: Yeah, this is, this is the like, the 1st time I’ve I’ve seen it working. So yeah, it’s pretty cool. It’s pretty amazing.

190 00:22:08.610 00:22:09.000 joshuadeveyra: And we’re all

191 00:22:09.530 00:22:11.340 joshuadeveyra: like self hosted right with them.

192 00:22:11.540 00:22:23.620 Uttam Kumaran: Yeah. So so a couple of things. One is, if anyone wants access to any of this, please just ping in the AI Channel. Miguel can send invites. We’re currently basically not paying for

193 00:22:23.830 00:22:32.119 Uttam Kumaran: anything like N. 8. And we have on the free tier and super base. We’re on the free tier drive, I guess. Drive we pay for. But that’s whatever. So

194 00:22:32.170 00:22:46.199 Uttam Kumaran: one of the things that we’re gonna end up doing, I mean hopefully, by the end of the year is we want to basically host like a workshop within the company where everybody can think of something that’s like a pain point for them that could be personally or professionally, and then we all walk through

195 00:22:46.340 00:22:50.913 Uttam Kumaran: like building it with one of these AI tools. But what I realized is

196 00:22:51.630 00:23:12.360 Uttam Kumaran: even for me, like I can see that these Demos, and see that there’s a lot of stuff going on and see that it’s possible. But until you go and touch and like and actually interact with these, it’ll give you a much better understanding of like what’s possible. And I want everybody in all capacities in the company to to know that we are going to use AI in their roles. And so one thing that we’ll try to host

197 00:23:12.664 00:23:26.069 Uttam Kumaran: as we get more confident is, isn’t that sort of workshop where everybody can log in and and create a workflow or an automation or anything? And again, I don’t care if it’s work related or personal, related. I just want to see

198 00:23:26.100 00:23:39.750 Uttam Kumaran: people try this stuff out. And I’m happy to to support that so could look forward to that at some point when we get some time. But yeah, everything here on N, 8 end is basically like it’s like a drag and drop

199 00:23:39.790 00:23:51.675 Uttam Kumaran: gui. It’s it’s more technical than I’m I’m explaining. But it’s very, very nice in terms of like building these these workflows. I’ll say kudos to Miguel.

200 00:23:52.280 00:24:01.979 Uttam Kumaran: you know he was, he told me. He’s like, Yeah, it looks like very complicated. And I’m like, well, your title is AI engineer. So you are an engineer now, and so it’s no longer like relying on

201 00:24:02.000 00:24:02.695 Uttam Kumaran: like

202 00:24:03.550 00:24:12.830 Uttam Kumaran: simple zapier like drag and drop stuff like this is the real deal. And I, it’s really amazing to kind of see progress. And this is just a few days.

203 00:24:13.390 00:24:36.470 Uttam Kumaran: so that’s just to say, like, you know, we have a lot of folks with traditional software backgrounds. Learning this stuff. We have people with mostly automation backgrounds going the other way. And so I’m really glad to kind of see our progress. We are going to always bias towards using open source tools because it’s cheaper. It’s more flexible. And it’s easier for us to navigate data security with clients.

204 00:24:36.520 00:24:45.680 Uttam Kumaran: So anytime where we can use these tools, we can. I I would prefer to so we’re we’re all finding out what’s where, what’s the right stack for us on this side as well, so.

205 00:24:47.420 00:24:50.269 Luke Daque: I wonder if this can do what we are

206 00:24:52.120 00:24:53.140 Luke Daque: workflows, though.

207 00:24:53.490 00:24:57.380 Uttam Kumaran: For sure, and it may be it may be even easier to develop, since it’s more visual.

208 00:24:58.080 00:24:58.760 Luke Daque: Yeah.

209 00:25:00.830 00:25:01.400 Luke Daque: cool.

210 00:25:01.400 00:25:10.540 Uttam Kumaran: But yeah, play around again. Anything we try, it’s gonna take iterations. It’s gonna take 2 or 3 or 4 tries with multiple different tools. Yeah, look, they already have.

211 00:25:10.540 00:25:11.340 Luke Daque: Yeah.

212 00:25:11.580 00:25:12.760 Luke Daque: exactly. Yeah.

213 00:25:12.760 00:25:13.360 Nicolas Sucari: You mean.

214 00:25:13.360 00:25:13.930 Luke Daque: Okay.

215 00:25:14.750 00:25:18.000 Nicolas Sucari: You mean, like the code review stuff that we were wanting to do with chat gpt.

216 00:25:18.220 00:25:19.280 Luke Daque: Yeah.

217 00:25:19.330 00:25:20.610 Luke Daque: or any other

218 00:25:21.540 00:25:22.780 Luke Daque: workflows.

219 00:25:24.540 00:25:25.370 Nicolas Sucari: Yeah, maybe that.

220 00:25:25.370 00:25:29.009 Luke Daque: So on on. Pull request. Yeah, cool. Nice.

221 00:25:29.130 00:25:30.359 Luke Daque: It looks like.

222 00:25:30.850 00:25:31.960 Luke Daque: looks like it.

223 00:25:33.390 00:25:34.010 Nicolas Sucari: Right.

224 00:25:34.880 00:25:35.880 Luke Daque: That’s amazing.

225 00:25:38.080 00:25:41.779 Uttam Kumaran: Cool. So that’s some of the stuff we’re doing on the AI side.

226 00:25:43.410 00:25:55.760 Uttam Kumaran: on the data side. I don’t know if there’s anything I mean, I think the only thing is pat on the data platform stuff. I know I don’t know if there’s anything. So say we can share on that, or if you want to, maybe

227 00:25:55.850 00:26:02.089 Uttam Kumaran: like I don’t know, or or we can, we can continue to wait until we have something like very concrete there. But.

228 00:26:02.512 00:26:12.237 Patrick Trainer: Actually keep talking. Give me a sec, and I’ll give you a concrete answer of if I wanna share something yet or not.

229 00:26:13.090 00:26:25.116 Uttam Kumaran: I other than that we have. The one thing I will ask and of course, like, do this if you want to, but we have some amazing content on Linkedin and on

230 00:26:26.680 00:26:36.810 Uttam Kumaran: On twitter and on tiktok now, so if you are obliged at all to go and like that, and repost and

231 00:26:37.613 00:26:42.689 Uttam Kumaran: the the number. One thing you could do is repost and add a comment about how you feel about it.

232 00:26:42.967 00:26:53.142 Uttam Kumaran: But would be very nice to kind of show people how to do that. If you go to our brain, forge Page, if you’re not following, please follow you can go to one of our posts.

233 00:26:55.240 00:26:58.929 Uttam Kumaran: one of our posts here, and then you can actually

234 00:26:58.970 00:27:01.390 Uttam Kumaran: go ahead and you can just click repost.

235 00:27:01.570 00:27:04.610 Uttam Kumaran: It’ll give you the option to repost with your thoughts.

236 00:27:06.140 00:27:21.390 Uttam Kumaran: this is gonna be the way we get money in the door among a bunch of different ways. But one of the things that I talked to the team about was right. Now we are a hundred percent coming from word of mouth or referrals.

237 00:27:21.470 00:27:45.330 Uttam Kumaran: which is good because I spent no money on sales for a long time. It’s bad because it’s very unpredictable. We may have months where we have nothing, we have months where there are people. And so we’re moving to a multi channel strategy, we’re gonna have content going out. We’re gonna have folks like Roy on the team who are direct dialing and interacting with clients, qualifying leads. We have emails going out from lead assassin.

238 00:27:45.656 00:28:06.520 Uttam Kumaran: So there’s a we’re gonna we want to have this variety of strategies going on. Content is one of the great ways to put our message out there. Allow people to really learn about us, and then, when the time is right, they can interact with us. And so we have stuff now on all platforms, and we just cut up the Github actions

239 00:28:06.700 00:28:31.810 Uttam Kumaran: talk. I did a while and put it on Tiktok. But we’ll start to put stuff there, too. We’re not looking for crazy numbers here. We’re mainly just looking for folks to see us know that we exist. Follow us. And you know there’s plenty of companies that I follow, that maybe it takes months for me to be like, okay, let me go buy a shirt from them or let me go message them and see what they’re really about. So this is, gonna take a long time. But it’s gonna be. It’s 1 i want us to be in the habit of

240 00:28:31.840 00:28:35.540 Uttam Kumaran: publishing. And ideally, I want everybody here to be able to publish

241 00:28:35.550 00:28:40.750 Uttam Kumaran: on the brand and build your own personal brand right? So there’s a ton of opportunity to do that as well.

242 00:28:40.830 00:28:59.109 Uttam Kumaran: In fact, I’ll probably that’ll be. The one thing I force everybody to do is to post some piece of content that you write, or at least we can talk, and then Ryan can take that and write a post about, because it helps to kind of narrate what you do here, and I think it’s helpful for everybody’s personal brand as well, so

243 00:29:01.420 00:29:02.210 Uttam Kumaran: cool.

244 00:29:02.840 00:29:03.980 Uttam Kumaran: Are you ready.

245 00:29:04.541 00:29:07.669 Patrick Trainer: Sorry I’m having to create a a token.

246 00:29:08.160 00:29:09.080 Patrick Trainer: but.

247 00:29:09.911 00:29:12.798 Uttam Kumaran: I’m trying to think anything else.

248 00:29:13.280 00:29:14.590 Patrick Trainer: Down, accrue.

249 00:29:14.590 00:29:21.630 Uttam Kumaran: We talked a bit about. We talked about the website. I guess I can share, I guess while we’re here I’ll just I’ll just share. Show off a little bit of what we’ve done

250 00:29:22.050 00:29:23.719 Uttam Kumaran: on the site.

251 00:29:26.050 00:29:29.650 Luke Daque: Yeah, that that was a good win as well for this week, right? This site.

252 00:29:30.230 00:29:31.730 Uttam Kumaran: Yeah, yeah.

253 00:29:32.370 00:29:34.740 Nicolas Sucari: The block point. The block point is ready.

254 00:29:35.520 00:29:41.669 Uttam Kumaran: We have all of our style guides. So for the folks on the engineering side, you can basically think of this as like

255 00:29:41.930 00:29:43.212 Uttam Kumaran: this is like,

256 00:29:43.780 00:29:44.820 Uttam Kumaran: documentation.

257 00:29:44.820 00:29:46.480 Nicolas Sucari: Components, yeah.

258 00:29:46.480 00:29:47.280 Uttam Kumaran: Like this is like.

259 00:29:47.280 00:29:47.790 Luke Daque: But.

260 00:29:48.340 00:30:03.680 Uttam Kumaran: This is like modules and like imports for basically for design. So and has gone like absolutely crazy here. And it’s amazing like, this is so nice to have, because we’ll basically be able to have all of this. And so the speed at which we can publish new pages

261 00:30:04.139 00:30:10.079 Uttam Kumaran: and publish new content. Is amazing. So we have everything on all of our major components here

262 00:30:10.190 00:30:15.679 Uttam Kumaran: that way. When we go to do redesigns and things like that we have. We have a, we have a design and a brand book.

263 00:30:16.020 00:30:21.779 Uttam Kumaran: You can see that we have like an update to the services page that we’re working on here.

264 00:30:22.080 00:30:26.129 Uttam Kumaran: We are also working on this new about us. Page

265 00:30:26.700 00:30:28.580 Uttam Kumaran: that will go live.

266 00:30:29.003 00:30:38.609 Uttam Kumaran: Again. We’re we’re working on like version 3 about of the site. Basically, we have our our blog like home page. Now, too. That’s that will go live.

267 00:30:38.770 00:30:45.929 Uttam Kumaran: you can see, like all this. You’ll be able to select the what news? Like, what category? And things like that.

268 00:30:46.210 00:30:48.684 Uttam Kumaran: So yeah, super excited here.

269 00:30:49.520 00:30:54.780 Uttam Kumaran: I guess. Erickson, maybe you if you want to give any update from your side as well.

270 00:30:55.316 00:31:03.039 Uttam Kumaran: That could be helpful. Again. I appreciate you joining these calls. So I wanna make sure everybody hears the the work that’s going on on your side.

271 00:31:04.330 00:31:04.910 Uttam Kumaran: Yep.

272 00:31:05.226 00:31:09.979 Ericson Dalusong: Thank you. So I’m just going to share my screen with you so I can

273 00:31:10.647 00:31:14.580 Ericson Dalusong: show you the the things that I’ve been working on for

274 00:31:14.968 00:31:19.419 Ericson Dalusong: for the past couple of weeks. So can you guys confirm if you can see it?

275 00:31:21.020 00:31:22.010 Ericson Dalusong: Yes.

276 00:31:22.200 00:31:33.379 Ericson Dalusong: okay, perfect. So basically, we’ve started to watch our outbound campaigns. Oops one second I’m showing the wrong screen.

277 00:31:40.040 00:31:44.260 Ericson Dalusong: Sorry I I just have a lot of tabs here.

278 00:31:46.200 00:31:47.370 Patrick Trainer: E, 2.

279 00:31:48.140 00:31:48.525 Ericson Dalusong: The

280 00:31:49.240 00:31:55.820 Ericson Dalusong: So there we go. So we started to launch our outbound campaigns targeting.

281 00:31:56.498 00:31:59.340 Ericson Dalusong: Companies in the e-commerce sector.

282 00:31:59.400 00:32:06.739 Ericson Dalusong: We’re also, you know, targeting companies that are dopa like of Stella Source. That’s 1 of the

283 00:32:07.090 00:32:16.450 Ericson Dalusong: the the clients that we have at Brainforge. And then we’ve also been targeting manufacturing companies. So these campaigns have already been launched.

284 00:32:16.490 00:32:34.300 Ericson Dalusong: And then just today, we were able to set up a couple of more campaigns the 1st one is, we work look alike. So we are going to target similar companies like we work.

285 00:32:34.400 00:32:38.040 Ericson Dalusong: And then we were also able to

286 00:32:38.606 00:32:39.699 Ericson Dalusong: to pull

287 00:32:40.160 00:32:45.160 Ericson Dalusong: all the companies that are hiring on. Indeed! So this

288 00:32:45.340 00:32:48.290 Ericson Dalusong: a high level view of the list that

289 00:32:48.984 00:32:52.459 Ericson Dalusong: list of companies that we’re going to target.

290 00:32:52.490 00:32:55.039 Ericson Dalusong: So these are all the positions that

291 00:32:56.770 00:32:58.690 Ericson Dalusong: we were able to scrape

292 00:32:59.266 00:33:02.949 Ericson Dalusong: on, indeed, using a tool called appify.

293 00:33:03.120 00:33:09.360 Ericson Dalusong: So what we’ve done here is that, you know, we’ve basically extracted the

294 00:33:10.062 00:33:12.900 Ericson Dalusong: the data, qualify them. And then

295 00:33:12.950 00:33:18.640 Ericson Dalusong: we we run an enrichment we’re in. We’re

296 00:33:18.940 00:33:24.039 Ericson Dalusong: we’re looking for the the contacts or the decision makers from this companies and from there

297 00:33:24.250 00:33:29.189 Ericson Dalusong: we’re going to to target them on cold email and and Linkedin.

298 00:33:29.210 00:33:30.340 Ericson Dalusong: So

299 00:33:30.480 00:33:40.219 Ericson Dalusong: these are the things that my team and I have been working on. We’ve only been running our campaign for for one week, and next week

300 00:33:40.240 00:33:47.710 Ericson Dalusong: we will be ramping up our volume. So that means that we’re gonna be targeting more people.

301 00:33:47.780 00:33:52.180 Ericson Dalusong: We’re gonna be sending more messages. And I’m very positive that

302 00:33:52.340 00:33:55.370 Ericson Dalusong: you know, we are going to get

303 00:33:55.530 00:33:59.550 Ericson Dalusong: positive responses and qualified meetings for the inforge.

304 00:34:00.680 00:34:08.449 Uttam Kumaran: Yeah, the only thing, the only thing here. So one thing that, Miguel we we worked on before is we have for the, for the finding people at the company

305 00:34:08.620 00:34:13.599 Uttam Kumaran: we we were started looking for. Actually the Hr people also to target

306 00:34:14.019 00:34:18.940 Uttam Kumaran: really finding out. Like, if there’s an Hr manager, someone where we basically say, Hey, we have

307 00:34:19.020 00:34:32.060 Uttam Kumaran: qualified leads, or we have we? We don’t. We have qualified expertise in this domain as well. We noticed that you were hiring cause that sometimes they’re the ones that are in control of actually the hiring process.

308 00:34:32.120 00:34:36.810 Uttam Kumaran: So I wonder if that’s if those are the people that when you go to do the find people at company step

309 00:34:37.335 00:34:40.314 Uttam Kumaran: maybe we can use those roles or

310 00:34:41.389 00:34:45.080 Uttam Kumaran: I I assume you must be. Yeah, I assume you’re going after the

311 00:34:45.360 00:34:53.490 Uttam Kumaran: these titles, or maybe we can just go after the like Hr managers and things like that. I can. And I can send you. I can send you an Apollo.

312 00:34:53.510 00:34:55.510 Uttam Kumaran: The process we set up before

313 00:34:56.060 00:34:56.980 Uttam Kumaran: Yup Yup.

314 00:34:57.271 00:35:03.469 Ericson Dalusong: That would be helpful. And the titles that I’ve added here with them and team just to share with you

315 00:35:03.689 00:35:06.241 Ericson Dalusong: are a combination of

316 00:35:06.989 00:35:09.849 Ericson Dalusong: you know, high level decision makers, because

317 00:35:10.009 00:35:15.569 Ericson Dalusong: some of these are, you know, probably, you know, working with with

318 00:35:15.929 00:35:18.069 Ericson Dalusong: with this decision makers.

319 00:35:18.469 00:35:25.539 Ericson Dalusong: And we’re also finding the the Hr managers. And you know, people in the recruitment tool

320 00:35:26.319 00:35:30.359 Ericson Dalusong: to target on our outbound campaigns for our outbound campaigns.

321 00:35:30.360 00:35:30.910 Uttam Kumaran: Okay.

322 00:35:31.400 00:35:32.840 Uttam Kumaran: cool. Yeah. But.

323 00:35:32.840 00:35:41.369 Ericson Dalusong: Yeah, I will wait for. I look forward to to getting the the list of titles. So I can also, you know, make some.

324 00:35:41.500 00:35:44.460 Ericson Dalusong: you know, refinements on our targeting.

325 00:35:46.240 00:35:46.990 Uttam Kumaran: Thank you.

326 00:35:47.500 00:35:48.729 Uttam Kumaran: Yup! Yup!

327 00:35:48.730 00:35:52.455 Ericson Dalusong: So yeah, that’s just it on on our side.

328 00:35:53.010 00:35:57.519 Ericson Dalusong: if you guys have any questions or feedback, just just let me know.

329 00:36:01.560 00:36:04.089 Patrick Trainer: Alright. So I’ve I’ve got everything set up

330 00:36:05.326 00:36:14.950 Patrick Trainer: it’s had to regenerate some Api keys and whatnot. So basically, what we’ve been talking about in terms of like data platform

331 00:36:15.030 00:36:26.190 Patrick Trainer: is kind of like a centralized control plane to have kind of like an overarching view of everything that’s going on in our system. So think.

332 00:36:26.200 00:36:29.209 Patrick Trainer: how do our Github actions run?

333 00:36:29.320 00:36:42.440 Patrick Trainer: How like are they failing? What assets and snowflake do we have? Are our 5 tran pipelines like on schedule, basically like health and like those sorts of things.

334 00:36:42.740 00:36:45.680 Patrick Trainer: And so I’ve started to

335 00:36:45.960 00:36:57.070 Patrick Trainer: built this out and so it’s at a pretty bare bones. State right now. Haven’t hooked like a a front end to it yet.

336 00:36:57.310 00:37:01.439 Patrick Trainer: but we should.

337 00:37:01.470 00:37:08.989 Patrick Trainer: That should be coming soon and so what I will show you is, let’s

338 00:37:10.090 00:37:12.479 Patrick Trainer: share this.

339 00:37:15.090 00:37:18.220 Patrick Trainer: Okay, hold on. I gotta open system settings.

340 00:37:22.930 00:37:26.336 Patrick Trainer: Okay, I I have to quit and reopen. Zoom, I’ll be back.

341 00:37:32.400 00:37:40.499 Uttam Kumaran: I know we’re going a little bit over. But I just want to share this last piece, because this is kind of like what Patrick has been spending a lot of time working on. So

342 00:37:48.680 00:37:49.440 Uttam Kumaran: okay.

343 00:37:49.830 00:37:50.570 Uttam Kumaran: cool.

344 00:37:50.570 00:37:52.220 Patrick Trainer: So let’s put this

345 00:37:52.950 00:37:54.410 Patrick Trainer: over here.

346 00:37:59.720 00:38:01.240 Patrick Trainer: where’d it go?

347 00:38:02.870 00:38:04.630 Patrick Trainer: Okay?

348 00:38:04.830 00:38:05.870 Patrick Trainer: Sure.

349 00:38:09.040 00:38:09.890 Patrick Trainer: Sure.

350 00:38:13.170 00:38:14.780 Patrick Trainer: Okay, you can see this.

351 00:38:17.350 00:38:21.319 Patrick Trainer: So basically, what we’ve got here is

352 00:38:22.200 00:38:24.980 Patrick Trainer: ignore. That is

353 00:38:24.990 00:38:38.160 Patrick Trainer: the I’ve wrote the the Api to be able to do this. And so what I’m going to demo is like just Github. Action runs. And so in this I just have, like a

354 00:38:38.220 00:38:48.869 Patrick Trainer: one of my repos that has a bunch of actions running constantly, and we want to be able to see, like the basically the

355 00:38:49.040 00:38:58.309 Patrick Trainer: status of what’s running in here. And so what I’ve written is this, like Api here

356 00:38:58.360 00:39:01.459 Patrick Trainer: that has it’s got a

357 00:39:02.210 00:39:06.309 Patrick Trainer: server, and then it’s everything’s in a plugin.

358 00:39:06.400 00:39:08.970 Patrick Trainer: And so we have this Github actions, Plugin.

359 00:39:09.050 00:39:11.510 Patrick Trainer: that is, calling

360 00:39:11.740 00:39:18.769 Patrick Trainer: the Api through this, and then looking for workflows and getting the health status for that.

361 00:39:18.810 00:39:24.560 Patrick Trainer: And so to do that we can do is we can start up the Api there

362 00:39:25.573 00:39:28.106 Patrick Trainer: and then we can.

363 00:39:28.800 00:39:30.590 Patrick Trainer: So this is the endpoint

364 00:39:30.690 00:39:34.049 Patrick Trainer: that it’s on. It’s serving here.

365 00:39:34.260 00:39:38.729 Patrick Trainer: And then we can actually, we can call the Api here.

366 00:39:38.910 00:39:41.190 Patrick Trainer: And so we have

367 00:39:42.810 00:39:48.120 Patrick Trainer: these different endpoints like we’ve got a metrics endpoint a health endpoint

368 00:39:48.556 00:39:59.639 Patrick Trainer: a metrics with the specific Plugin name, for when we have others, and then, like health endpoint with with a plugin name. And that’s just so we can be specific.

369 00:39:59.780 00:40:05.310 Patrick Trainer: The plugin name does get printed out like here

370 00:40:05.849 00:40:12.840 Patrick Trainer: that’s where we’re at. And so, but like what we can do is like we can hit the health endpoint.

371 00:40:13.400 00:40:15.920 Patrick Trainer: And that’s gonna hit. And it’s gonna

372 00:40:16.060 00:40:20.510 Patrick Trainer: basically what it’s doing. It’s making a request to Github

373 00:40:20.520 00:40:33.630 Patrick Trainer: and seeing if all of that is good, if it’s basically alive. So I think if we have, like like Snowflake, or some server running, and we just wanna see if it’s alive. We can use this to hit it, hit it there

374 00:40:34.176 00:40:42.389 Patrick Trainer: and then, if we also want to see like metrics. But we want to see it for our

375 00:40:42.820 00:40:43.860 Patrick Trainer: Github

376 00:40:44.330 00:40:45.680 Patrick Trainer: actions.

377 00:40:47.160 00:40:51.540 Patrick Trainer: and then we can actually pipe this into Jq, we can hit that.

378 00:40:52.020 00:40:53.010 Patrick Trainer: And

379 00:40:53.260 00:40:54.969 Patrick Trainer: it’s going to return

380 00:40:55.470 00:40:57.150 Patrick Trainer: the data

381 00:40:57.240 00:40:59.600 Patrick Trainer: of our workflow

382 00:40:59.820 00:41:02.140 Patrick Trainer: and the status that it’s done.

383 00:41:02.430 00:41:04.180 Patrick Trainer: And so what this

384 00:41:04.750 00:41:11.169 Patrick Trainer: then, corresponds to is like all of these here. So you see, we have this run.

385 00:41:11.410 00:41:13.869 Patrick Trainer: we have a run. Id there.

386 00:41:18.010 00:41:19.730 Patrick Trainer: Think it’s actually going to be

387 00:41:21.170 00:41:21.950 Patrick Trainer: cool

388 00:41:24.980 00:41:26.140 Patrick Trainer: runs.

389 00:41:27.570 00:41:29.679 Patrick Trainer: Okay? Yeah. I think that’s it.

390 00:41:29.990 00:41:31.430 Patrick Trainer: Well, anyway.

391 00:41:32.620 00:41:34.439 Patrick Trainer: That id

392 00:41:34.940 00:41:36.960 Patrick Trainer: goes to something. But

393 00:41:37.380 00:42:00.139 Patrick Trainer: if you I know this is just like an Api. But if you think are very bare bones, Json, but if you think about what a front end would look like on this, you could have like a page, and then, like a like a little box in the middle with that, then has, like a list of your different pipelines that have ran.

394 00:42:02.650 00:42:04.280 Patrick Trainer: yeah, that’s pretty much it.

395 00:42:07.020 00:42:14.229 Uttam Kumaran: Yeah, what we’re basically thinking about is like as we span across multiple clients, we’re just have a lot of stuff running in the background.

396 00:42:14.794 00:42:38.415 Uttam Kumaran: And how do you centralize all that information and basically give yourself like a health of the client from the data side. And this is not only just going to be like Github. It’s going to be like snowflake. Dbt, like 5 train. We can start to pipe in a lot of metrics. But again, basically, what this should do and how this actually impact the client is this should reduce the number of

397 00:42:38.920 00:42:55.812 Uttam Kumaran: like this should basically reduce the number of bugs they should improve how fast we’re able to fix those issues. And then the last thing this should basically help us help our engineers internally, you know, focus on the right stuff. So that’s kind of like what this whole

398 00:42:56.150 00:42:56.840 Patrick Trainer: Okay, so.

399 00:42:56.840 00:42:59.199 Uttam Kumaran: So the mission on the platform side is for.

400 00:42:59.450 00:43:04.390 Patrick Trainer: So here I have a this other repo here

401 00:43:04.630 00:43:10.189 Patrick Trainer: which would actually have some failures in it. So let’s go over to this M. File

402 00:43:14.090 00:43:15.580 Patrick Trainer: actions.

403 00:43:18.410 00:43:19.240 Patrick Trainer: Oh.

404 00:43:22.610 00:43:24.669 Patrick Trainer: and then let’s try and

405 00:43:26.560 00:43:28.000 Patrick Trainer: call it.

406 00:43:28.400 00:43:31.359 Patrick Trainer: I’m gonna have to restart the server.

407 00:43:33.390 00:43:34.190 Patrick Trainer: No.

408 00:43:35.570 00:43:36.980 Patrick Trainer: let’s call it again.

409 00:43:47.540 00:43:49.974 Patrick Trainer: I don’t know what happened, but

410 00:43:50.380 00:43:54.100 Nicolas Sucari: To change the name of the repo on the

411 00:43:54.560 00:43:56.140 Nicolas Sucari: second terminal, I think.

412 00:43:56.160 00:43:57.219 Nicolas Sucari: and you have there.

413 00:43:59.640 00:44:02.989 Nicolas Sucari: It’s not Github action, the metrics, one after metrics. You don’t.

414 00:44:02.990 00:44:05.919 Patrick Trainer: Oh, yeah, that’s that’s the endpoint.

415 00:44:06.230 00:44:07.430 Patrick Trainer: But

416 00:44:12.460 00:44:14.519 Patrick Trainer: yeah, that that shouldn’t matter

417 00:44:19.990 00:44:20.970 Patrick Trainer: anyway.

418 00:44:21.570 00:44:25.099 Patrick Trainer: Still, a work in work in progress where? All right? Oh.

419 00:44:25.220 00:44:26.333 Patrick Trainer: wait! Oh.

420 00:44:27.130 00:44:35.590 Nicolas Sucari: So this is only for Github right now. So which one you’re talking about using this for Dbt. Snowflake, Fivetran, and all of the other ones, and

421 00:44:35.760 00:44:48.040 Nicolas Sucari: how you’re gonna like you’re gonna by, through an Api like request different things for to those tools so that we can get like that responses and see if they have. If they are working fine.

422 00:44:50.270 00:44:51.910 Nicolas Sucari: or or how it’s like

423 00:44:52.340 00:44:55.889 Nicolas Sucari: what you’re gonna like. Try to do the same with Snowflake.

424 00:44:55.990 00:44:57.040 Nicolas Sucari: for example.

425 00:44:58.380 00:45:07.309 Uttam Kumaran: Yeah, so you can think of it as like, basically whenever we interact with snowflake or any of our data tools, we basically want to have a single layer where we do the interaction.

426 00:45:07.380 00:45:11.479 Uttam Kumaran: whether it’s like spinning up instances, modifying things like that.

427 00:45:13.620 00:45:14.520 Uttam Kumaran: But

428 00:45:14.660 00:45:19.659 Uttam Kumaran: but yeah, basically, I think Patrick is like a longer roadmap about kind of how the stages in which we do this.

429 00:45:19.750 00:45:41.170 Uttam Kumaran: But we want to start with just like bringing in all the data from each of those sources, and then also like creating databases, creating schemas like all that stuff, it’s really redundant work. And it’s likely work that we can at least print like Templatize, and then, at most, I think we’ll probably end up just building agents that start to take care of some of that stuff for us.

430 00:45:43.380 00:45:46.530 Luke Daque: Nice and is the end goal here to like create

431 00:45:46.830 00:45:49.169 Luke Daque: a web page or something where we.

432 00:45:49.170 00:45:49.720 Patrick Trainer: More or less.

433 00:45:49.720 00:45:51.170 Luke Daque: That’s all the errors the hell.

434 00:45:51.170 00:45:56.810 Patrick Trainer: Yeah, it’d be. It’d be like like an admin page where you get like a a holistic view

435 00:45:56.910 00:45:58.030 Patrick Trainer: of

436 00:45:58.170 00:46:00.402 Patrick Trainer: all of our assets.

437 00:46:01.010 00:46:06.440 Patrick Trainer: all of our yeah, everything that’s running so that we can like easier

438 00:46:06.560 00:46:18.276 Patrick Trainer: or having much easier time managing all of the stuff that we need to look at and like this gets especially complex, like once we have like multiple clients.

439 00:46:18.680 00:46:19.010 Luke Daque: Right.

440 00:46:19.010 00:46:28.890 Patrick Trainer: And so then, like this is what would allow you to like dynamically check what’s going on, instead of like having to go to a bunch of different places and do it.

441 00:46:29.320 00:46:29.910 Uttam Kumaran: Yeah.

442 00:46:29.910 00:46:32.600 Luke Daque: Yeah, that’s that’s awesome. This is awesome.

443 00:46:33.260 00:46:35.139 Luke Daque: Yeah, we need Patrick.

444 00:46:35.140 00:46:41.760 Uttam Kumaran: The platform supports all the engineers that are on it. Right? So it supports like Nico, who has to communicate. If there’s issues

445 00:46:42.460 00:46:47.480 Uttam Kumaran: also, that becomes something that we we want to sell and we want to have is the way we do this work, you know. So

446 00:46:48.110 00:46:55.710 Uttam Kumaran: like, this is the sort of R&D that I’m like, very, very excited that we can have. And then basically help us with our clients. So

447 00:46:58.430 00:47:04.528 Uttam Kumaran: cool. Okay, I know we’re way over. But any other questions or anything else I can

448 00:47:05.600 00:47:06.850 Uttam Kumaran: answer.

449 00:47:12.080 00:47:20.650 Uttam Kumaran: Cool. Well, if not, thanks, guys, this is a great week. I’m gonna be out next week. Nico’s gonna be running these meetings Monday, Friday.

450 00:47:21.017 00:47:25.029 Uttam Kumaran: I’ll still be on slack here and there, but I think everybody kind of has

451 00:47:25.452 00:47:28.799 Uttam Kumaran: all their stuff for next week. But just

452 00:47:29.080 00:47:31.364 Uttam Kumaran: please let me know if there’s anything urgent.

453 00:47:32.890 00:47:33.920 Uttam Kumaran: but cool.

454 00:47:34.570 00:47:35.460 joshuadeveyra: Thanks. Everyone.

455 00:47:35.740 00:47:36.660 Uttam Kumaran: Thanks guys.

456 00:47:36.870 00:47:38.140 Luke Daque: Have a nice rest of your day.

457 00:47:38.140 00:47:38.450 Ericson Dalusong: Thank you.

458 00:47:38.450 00:47:39.170 Nicolas Sucari: Guys we get.

459 00:47:39.170 00:47:39.820 Roy Christian Piñon: Guys.

460 00:47:40.380 00:47:41.349 Luke Daque: Bye, bye.