Meeting Title: Eden Project Check-in and Feedback Date: 2025-09-05 Meeting participants: Henry Zhao, Uttam Kumaran
WEBVTT
1 00:00:26.690 ⇒ 00:00:27.740 Henry Zhao: Well, that was fast.
2 00:00:27.900 ⇒ 00:00:28.740 Uttam Kumaran: Yay.
3 00:00:30.120 ⇒ 00:00:31.259 Uttam Kumaran: How’s it going?
4 00:00:31.400 ⇒ 00:00:32.229 Henry Zhao: How are you doing?
5 00:00:32.580 ⇒ 00:00:33.480 Uttam Kumaran: Yeah, dude.
6 00:00:33.580 ⇒ 00:00:35.350 Uttam Kumaran: How’s everything this week?
7 00:00:35.350 ⇒ 00:00:36.710 Henry Zhao: Yeah, that was an intense week.
8 00:00:37.220 ⇒ 00:00:37.965 Uttam Kumaran: Yeah?
9 00:00:38.710 ⇒ 00:00:39.600 Henry Zhao: How’s it?
10 00:00:39.600 ⇒ 00:00:42.169 Uttam Kumaran: How’s it going with, with Eden’s stuff?
11 00:00:42.770 ⇒ 00:00:45.870 Henry Zhao: Really good! So, like, I picked up Tableau this week, obviously.
12 00:00:46.790 ⇒ 00:00:53.530 Henry Zhao: So, a lot of Tableau stuff, a lot of stuff for JUD in terms of customer I.O, a lot of stuff in terms of the tracking stuff.
13 00:00:53.830 ⇒ 00:01:00.150 Henry Zhao: Yeah, and then just also trying to hold down the fort without Robert available 24-7.
14 00:01:00.350 ⇒ 00:01:01.050 Uttam Kumaran: Yeah.
15 00:01:01.050 ⇒ 00:01:06.079 Henry Zhao: Like, yeah, I kind of want to get your sense of Eden, and, like, we made…
16 00:01:06.080 ⇒ 00:01:10.789 Uttam Kumaran: I mean, on the data team, I think we’re bringing some new people in, we’re moving some people out, I think…
17 00:01:10.990 ⇒ 00:01:28.460 Uttam Kumaran: I think your background is amazing because you’ve done so much, and I think you’re good at, like, sort of being a chameleon for whatever is necessary, but I guess, like, this Tableau stuff is coming. I guess I talked to Awash a little bit about, sort of, the different scopes of work, like, how do you feel about the stuff you’re taking on?
18 00:01:28.460 ⇒ 00:01:36.029 Uttam Kumaran: for Eden, do you feel like it’s all uncomfortable? Are you, like, yo, we’re, like, basically underwater, or you’re like, oh, this is, like.
19 00:01:36.340 ⇒ 00:01:39.879 Uttam Kumaran: Cake? Like, give me a sense of, like, where on the spectrum you think we’re at.
20 00:01:39.880 ⇒ 00:01:49.439 Henry Zhao: Yeah, it’s super accomplishable. I think in the beginning, I felt more underwater, because I was onboarding, I felt like. I felt like I had… first of all, I had to get used to how intense Eden is.
21 00:01:49.440 ⇒ 00:01:50.910 Uttam Kumaran: Yes.
22 00:01:50.980 ⇒ 00:01:55.790 Henry Zhao: and build those relationships, so I think I’ve done that now, so it’s a lot more manageable right now.
23 00:01:55.950 ⇒ 00:02:03.040 Henry Zhao: Because I think in the beginning, it felt underwater because I had to manage the relationships, people were asking me questions where it’s like, I don’t know how to answer that yet.
24 00:02:03.230 ⇒ 00:02:10.710 Henry Zhao: So I was marking so many things in Slack as, like, remind me in 1 hour, remind me in 3 hours, just because I couldn’t keep up with, like, the inflow of questions.
25 00:02:10.910 ⇒ 00:02:11.510 Uttam Kumaran: Yeah.
26 00:02:11.510 ⇒ 00:02:19.720 Henry Zhao: It felt like I was drinking out of a fire hose, but now, like, I feel like that flow is manageable, and I’m, like, able to drink more quickly, if that makes sense.
27 00:02:19.720 ⇒ 00:02:20.340 Uttam Kumaran: Yes.
28 00:02:20.340 ⇒ 00:02:22.080 Henry Zhao: So it’s like, it’s still a fire hose, but I’m.
29 00:02:22.080 ⇒ 00:02:25.910 Uttam Kumaran: You’re able to consume it, you’re able to, like, yeah, yeah, nice, okay.
30 00:02:25.910 ⇒ 00:02:31.199 Henry Zhao: There’s, like, 50% of the questions now I can answer off the top of my head, whereas 50% I still need to look into.
31 00:02:31.360 ⇒ 00:02:37.409 Henry Zhao: Right? But that’s more manageable than when 100% of the things I need to look into while acting like I know what’s going on.
32 00:02:37.410 ⇒ 00:02:41.239 Uttam Kumaran: Because I don’t want it to seem like Robert Braun, somebody that, like, knows nothing, right? So…
33 00:02:41.500 ⇒ 00:02:50.129 Henry Zhao: that was the challenge in the beginning, now I think everything is doable, but I still need to get used to all the things that are in Tableau, like I need to understand how Annie built stuff.
34 00:02:50.670 ⇒ 00:02:57.250 Henry Zhao: I need to get used to working with Judd, and then now Devin, and Tracy, those new people that were brought on.
35 00:02:57.960 ⇒ 00:03:00.320 Henry Zhao: And yeah, just one step at a time.
36 00:03:01.000 ⇒ 00:03:06.509 Uttam Kumaran: like, I guess on the Tableau work, like, and this is maybe a question I had for Awash, too, like.
37 00:03:06.630 ⇒ 00:03:18.010 Uttam Kumaran: let me know if it’s… if it’s better for us to just bring in, like, someone really junior that all they do is, like, Tableau, and, like, maybe you need that, or if you’re like, hey, it’s actually helpful for me to take this on.
38 00:03:19.250 ⇒ 00:03:26.319 Uttam Kumaran: because we’ve always sort of, like… I think there’s, like, dashboarding, and then there’s, like, other parts of analysis. There’s a lot of people that can do the… do…
39 00:03:26.580 ⇒ 00:03:33.730 Uttam Kumaran: like, change a dimension title, add a new column, stuff like that. So if you think that, like, hey, there’s actually a good amount of work that is
40 00:03:33.840 ⇒ 00:03:37.210 Uttam Kumaran: Like, very, very basic dashboarding stuff.
41 00:03:37.470 ⇒ 00:03:41.180 Uttam Kumaran: I would be great to have, like, 5 or 10 hours with someone that all they do is that.
42 00:03:41.310 ⇒ 00:03:42.060 Uttam Kumaran: I do.
43 00:03:42.060 ⇒ 00:03:46.040 Henry Zhao: I don’t think that’s… I don’t think that’s an efficient use of company resources.
44 00:03:46.040 ⇒ 00:03:54.590 Uttam Kumaran: I think I’d rather just take it on, like, in the beginning, it’ll be a few more hours, because I need to understand the dashboards, understand where the data’s coming from, but after that, it should just be very low maintenance.
45 00:03:54.740 ⇒ 00:03:55.790 Uttam Kumaran: Okay, okay, cool.
46 00:03:55.790 ⇒ 00:03:56.320 Henry Zhao: Yeah.
47 00:03:56.550 ⇒ 00:03:59.670 Uttam Kumaran: And then tell me how you feel about, like, that plus default now.
48 00:04:01.190 ⇒ 00:04:13.050 Henry Zhao: I feel like default is still not that many hours for me, just because a lot of it is still on Vishal. It’s like me being able to communicate with him on what we need, and managing that time well. So, I think default is super manageable.
49 00:04:13.410 ⇒ 00:04:14.090 Uttam Kumaran: Okay.
50 00:04:14.340 ⇒ 00:04:20.380 Uttam Kumaran: And then, yeah, I guess, like, I would love to hear, sort of, like, your feedback on… just like…
51 00:04:20.529 ⇒ 00:04:38.499 Uttam Kumaran: how you see the company changing since you’ve been here, and yeah, I just want to kind of hear about, like, kind of get a sense for, like, what hours have been so far, and then, yeah, just hear, like, okay, kind of want to try to talk about where we can tap you in for more, or, like, how more you want to get involved, and yeah, just, like.
52 00:04:38.670 ⇒ 00:04:40.719 Uttam Kumaran: Just generally, you want to get a pulse check.
53 00:04:41.030 ⇒ 00:04:49.789 Henry Zhao: Yeah, I like that Robert and you are bringing in new clients, because I think once I get used to Eden, and, you know, default is further along.
54 00:04:49.960 ⇒ 00:04:53.289 Henry Zhao: There will be a lot of bandwidth to, you know, help out different clients.
55 00:04:53.500 ⇒ 00:04:55.799 Henry Zhao: Right? Otherwise, we’ll just be, like.
56 00:04:56.040 ⇒ 00:04:58.600 Uttam Kumaran: Kind of maintaining things, and… Yeah.
57 00:04:58.690 ⇒ 00:05:01.260 Henry Zhao: You know what I mean? Like, the incremental amount of…
58 00:05:01.760 ⇒ 00:05:11.910 Henry Zhao: like, in the future, as more things come in, the incremental amount of time that it’ll take will be much and much lower, right? So it’s good for us to continue to feed that pipeline and keep our teams lean.
59 00:05:12.030 ⇒ 00:05:18.579 Henry Zhao: Yeah. Especially in the current economy, I think… I think what happened during the pandemic was all of the companies were seeing growth.
60 00:05:18.870 ⇒ 00:05:23.430 Henry Zhao: Especially tech, so they started hiring all these different analysts and engineers, and…
61 00:05:23.850 ⇒ 00:05:28.279 Henry Zhao: Cross-functional managers, product managers, where it was, like, overinflated.
62 00:05:28.280 ⇒ 00:05:28.780 Uttam Kumaran: Yes.
63 00:05:28.780 ⇒ 00:05:32.979 Henry Zhao: Now it’s kind of in that, like, overcorrection. Now it’s like we’re correcting back to normal.
64 00:05:33.340 ⇒ 00:05:37.659 Henry Zhao: And I think lean teams and bringing on new clients is the way to go.
65 00:05:38.520 ⇒ 00:05:48.050 Uttam Kumaran: And then tell me, I guess, like, how interested are you in, like… so, I mean, I guess one thing that we’ve always been nervous about is, like, having people… I don’t think we’ve successfully stretched
66 00:05:48.680 ⇒ 00:05:55.309 Uttam Kumaran: Like, any of the engineers or people on engineering beyond, like, two clients? Usually, it’s tough.
67 00:05:55.530 ⇒ 00:06:07.580 Uttam Kumaran: Like, so I guess I’m… I’m curious, like, what you think about that. Also, like, curious how… if you have feedback about, like, our delivery overall, our project management, and, like, I think…
68 00:06:07.740 ⇒ 00:06:23.780 Uttam Kumaran: I like it because you’re a very decisive and opinionated person, but also, like, I think you’re on the ground, and you probably know a lot about what the day-to-days are like, and can give advice back on, like, how do we build this system of, you know, delivery?
69 00:06:23.800 ⇒ 00:06:39.860 Uttam Kumaran: But yeah, just, like, a couple of questions there on, like, what you think about stretching to a third client, or if you think it’s better of, like, instead, I would rather continue what I’m doing than maybe train someone up and then move on to the more complicated stuff. Like, kind of, like, what your…
70 00:06:39.950 ⇒ 00:06:43.559 Uttam Kumaran: What your goals are, or, like, what you would want to do, sort of, long-term.
71 00:06:44.070 ⇒ 00:06:48.509 Henry Zhao: Yeah, I’m not an engineer, so I can definitely stretch to more than two clients, but I do agree that.
72 00:06:48.510 ⇒ 00:06:59.350 Uttam Kumaran: You’re an engineer, dude, like, I think everybody on our development team is… we’re all doing engineering work, so I wouldn’t sell yourself short there, for sure.
73 00:07:00.320 ⇒ 00:07:03.589 Henry Zhao: Oh, no, not at all, but I think I can definitely stretch to more than 2 clients.
74 00:07:03.650 ⇒ 00:07:04.330 Uttam Kumaran: Okay.
75 00:07:04.410 ⇒ 00:07:09.420 Henry Zhao: Yeah, when I was, like, just, like, doing it on my own, I’ve had up to, like, 5 clients, and it was going pretty well.
76 00:07:09.770 ⇒ 00:07:17.330 Henry Zhao: And one of the benefits of that was I would learn some things from one or two clients that I can bring to other clients. So there was, like, economies of scale there, too.
77 00:07:17.540 ⇒ 00:07:21.420 Uttam Kumaran: Yeah, so that’s kind of, like, what we’re finding, and, like, one of the things that
78 00:07:21.490 ⇒ 00:07:38.930 Uttam Kumaran: like, Shreya’s working on these playbooks, and I think we’re going to start to do more of those. That is a huge help for us to start to think about, like, the services we’re offering and how to do that. But I mean, I don’t know, I guess, like, even beyond just client work, like, is that kind of where you want to stick? Do you feel like you want to go towards
79 00:07:39.310 ⇒ 00:07:52.829 Uttam Kumaran: a particular function, like, I guess trying to get a sense of, like, okay, if we were to map out, like, next 6 months, like, kind of, like, where do you see yourself fitting? Is it just kind of… kind of same-same, and just, like, pure development work, or, like…
80 00:07:52.900 ⇒ 00:08:03.779 Uttam Kumaran: for us, I’ve really enjoyed working with you. I think you also responded really well to our initial feedback and corrected, and it’s been really, really great. So for me, I think about, like.
81 00:08:03.950 ⇒ 00:08:17.739 Uttam Kumaran: pouring gasoline on people like you in a good way, where I want to sort of make sure that you’re in the spot you want it, and the company is gonna grow, and my interest is always going to be from promoting within, like.
82 00:08:17.810 ⇒ 00:08:24.149 Uttam Kumaran: moving people into higher leverage positions versus going and getting people from the outside. So.
83 00:08:24.540 ⇒ 00:08:31.390 Uttam Kumaran: They could be something you could think about, it could be something you could answer now, but yeah, just kind of, like, want to get a sense of, like, where you think this could go.
84 00:08:31.740 ⇒ 00:08:41.129 Henry Zhao: Yeah, I want to be that chameleon, I want to be able to do a little bit of everything, so just put me into whatever is needed, and I like figuring things out and filling those gaps.
85 00:08:41.130 ⇒ 00:08:43.649 Uttam Kumaran: How interested are you on the AI side of stuff?
86 00:08:44.540 ⇒ 00:08:47.359 Henry Zhao: Yeah, I want to learn, I just think I don’t have too much time.
87 00:08:47.360 ⇒ 00:08:51.430 Uttam Kumaran: That’s what I asked. I asked interest. I didn’t say, how much do you know? I said interest, interest.
88 00:08:52.020 ⇒ 00:08:53.609 Henry Zhao: But high, high interest, of course.
89 00:08:53.830 ⇒ 00:08:54.880 Uttam Kumaran: Okay, okay.
90 00:08:55.200 ⇒ 00:09:08.410 Uttam Kumaran: The reason I like… the reason is, one, nobody knows the AI stuff, it’s only been… it’s only been for, like, a year and a half, so don’t worry about, like, figuring it out. It’s actually pretty easy. Yeah. What’s hard is, like, the people stuff.
91 00:09:08.500 ⇒ 00:09:22.689 Uttam Kumaran: Where requirements gathering is really difficult when some of the clients we talk to don’t really know the systems, so there’s a lot of interaction with people. Second, it’s, a lot of testing. Like, think about, like.
92 00:09:22.800 ⇒ 00:09:40.340 Uttam Kumaran: what we’re using in ChatGBT, for example, is a general purpose thing, right? The complex thing just goes, like, do you have a nice prompt, and then you put something in. But imagine if you’re putting in unstructured, structured data, you then need it to go somewhere, we run evaluations, like, we use an LLM to judge the output.
93 00:09:40.340 ⇒ 00:09:45.940 Uttam Kumaran: across, like, a golden data set, so it actually gets a little bit more complicated, but I think in…
94 00:09:45.960 ⇒ 00:09:48.140 Uttam Kumaran: although I do love the data stuff.
95 00:09:48.420 ⇒ 00:09:55.619 Uttam Kumaran: I think as a data person in the company, if you’re interested in also, like, taking on some of the AI work, we have some really, really
96 00:09:55.960 ⇒ 00:10:01.480 Uttam Kumaran: really interesting AI work that we’re taking for some clients, that just knowing you, I feel like
97 00:10:01.720 ⇒ 00:10:07.300 Uttam Kumaran: you would… I think if you just saw, you would find a way, and it could start as just, like.
98 00:10:07.510 ⇒ 00:10:23.720 Uttam Kumaran: because a lot of the times for the AI clients, we struggle because the clients don’t help us a lot in, like, testing things really frequently. They don’t give us great feedback, so it’s almost like we need some of that support. Also, I think you’d be surprised at how
99 00:10:23.730 ⇒ 00:10:29.410 Uttam Kumaran: like, it’s actually not that complicated, but it is not like normal engineering. It’s like a…
100 00:10:29.810 ⇒ 00:10:34.090 Uttam Kumaran: It’s a lot of, like, prompt writing, it’s a lot of, like, optimizing outputs.
101 00:10:34.340 ⇒ 00:10:35.770 Uttam Kumaran: Yeah.
102 00:10:35.770 ⇒ 00:10:49.540 Henry Zhao: But I feel like it kind of has some of the same concepts as engineering, where you have to think of possible customer situations, like SQL injection, or whatever happens, right? Like, when you’re testing, you have to think of all these possible outcomes and solve for those, so…
103 00:10:49.540 ⇒ 00:10:50.370 Uttam Kumaran: Yeah.
104 00:10:50.370 ⇒ 00:10:51.640 Henry Zhao: It’s kind of related, yeah.
105 00:10:51.780 ⇒ 00:11:07.650 Uttam Kumaran: Okay, cool, so let me, let me talk to Sam, and then think about, like, where… I think if you, like… what has sort of been your, like, bandwidth these days? Like, how… I guess I haven’t… I didn’t get a chance to check out Clockify, but, like, how has bandwidth been these last, like, 3-4 weeks?
106 00:11:08.310 ⇒ 00:11:18.170 Henry Zhao: Yeah, so I’m, like, opening up bandwidth in terms of my life, but, like, for example, I felt like I worked 40 hours on Eden this week, but it was only, like, 14 so far, so, like…
107 00:11:18.910 ⇒ 00:11:21.579 Henry Zhao: So I don’t know if I just spent more time thinking about things than, like.
108 00:11:21.580 ⇒ 00:11:22.560 Uttam Kumaran: Yeah, yeah.
109 00:11:22.740 ⇒ 00:11:23.730 Henry Zhao: And sending stuff.
110 00:11:24.320 ⇒ 00:11:28.889 Henry Zhao: But yeah, I’m gonna continue to open up bandwidth, if there’s work to be done.
111 00:11:29.120 ⇒ 00:11:42.579 Uttam Kumaran: Yeah, so I guess that’s my question, too, like, how open are you to taking on more stuff, and, like, okay, you still feel comfortable with, kind of, like, the current hourly model? One of the things I would like to do for people, like.
112 00:11:42.640 ⇒ 00:11:54.129 Uttam Kumaran: we don’t have to do this anytime soon, but if it gets to a point where you’re kind of consistently doing a certain amount of hours, and it’s easier to move to something more fixed, we can do that as well.
113 00:11:54.130 ⇒ 00:12:02.679 Uttam Kumaran: I don’t mind either way. If anything, if we move to fixed, it would just give us some ability to kind of predict hours, but if you’re fine where you’re at, and
114 00:12:02.770 ⇒ 00:12:09.980 Uttam Kumaran: you, you know, you kind of indicate you have, like, an extra 10-15 hours, like, I’ll look to see, because I think your…
115 00:12:10.230 ⇒ 00:12:16.159 Uttam Kumaran: your background is really, really good. I think we could use your help in a couple different areas, in addition to data, but…
116 00:12:16.310 ⇒ 00:12:24.519 Uttam Kumaran: for me, I’m most interested in you helping us build, like, not only execute for clients, but also get your feedback on everything about how we deliver.
117 00:12:25.240 ⇒ 00:12:39.220 Uttam Kumaran: Okay. And so, continuing to give me feedback about the ways we replicate some of these solutions across clients, I would love… I’m super excited to see the next, like, playbook review, but those are the things where I think
118 00:12:41.260 ⇒ 00:12:51.450 Uttam Kumaran: I… some people can’t contribute in that way, like, they’re just clock in, clock out, but for something like you, I would like you to start to think of how you could help us sort of build this delivery engine.
119 00:12:51.500 ⇒ 00:13:02.919 Uttam Kumaran: You know, how we can do things efficiently, how… and most effectively, most importantly, like, communicate. Like, how can we communicate what the outcomes are, the time it’s gonna take, who’s gonna be involved for projects?
120 00:13:02.980 ⇒ 00:13:16.230 Uttam Kumaran: I’m telling you, if you do that, you can already get access to that as much as you want, but I would say, like, I want to put you in the positions where you’re motivated, and I can tell it’s, like, something new for you, so that you’ll start to get interested. The AI stuff.
121 00:13:16.260 ⇒ 00:13:32.229 Uttam Kumaran: is, like, really, really the coolest thing. I would say, I know you… you were sort of in a meeting with Mustafa on the go-to-market stuff for default, too. That sort of, like, dynamic data enrichment stuff is really, really cool. So anything like that that’s there, I’ll… I’ll start to… to loop you in.
122 00:13:32.490 ⇒ 00:13:33.250 Henry Zhao: Okay, cool.
123 00:13:33.960 ⇒ 00:13:38.989 Henry Zhao: And like I said, I think the first time we talked, I also don’t mind doing things that are, like, tedious or, like, repetitive.
124 00:13:38.990 ⇒ 00:13:39.320 Uttam Kumaran: Yeah.
125 00:13:39.540 ⇒ 00:13:41.370 Henry Zhao: There are benefits to that as well.
126 00:13:41.770 ⇒ 00:13:48.680 Henry Zhao: Throughout my career, the only thing I’ve ever not liked was what I said earlier, when, like, teams get too bloated. Yes.
127 00:13:49.060 ⇒ 00:13:55.139 Uttam Kumaran: Dude, but this is so nice for you to say that, because I… yes, I am in the same camp.
128 00:13:55.390 ⇒ 00:14:12.779 Uttam Kumaran: I… we… as I mentioned, like, I would totally promote from within and give people more responsibility than bring someone externally. Like, that’s the first place I go to, is the current crew, and now what I want to do, instead of bringing people externally at, like, a higher level, I’d rather just, like.
129 00:14:12.780 ⇒ 00:14:19.150 Uttam Kumaran: Build a crew of great leaders, functional people, and then slot people under you guys.
130 00:14:19.170 ⇒ 00:14:33.139 Uttam Kumaran: And naturally, it’ll sort of float people up, but I’m not interested in getting people that are, like, big consultant, bigwigs, super serious, like, boring PMs. Like, I kind of want people who are, like.
131 00:14:33.340 ⇒ 00:14:48.610 Uttam Kumaran: this could be, like, we don’t have to have a meeting about this, we don’t need to. Or, like, they challenge sort of a lot of that, because I don’t want our team in meetings, and I want to pay people more competitively, which means the business has to make more, and revenue per employee has to be high for me to start to pay
132 00:14:48.720 ⇒ 00:14:52.089 Uttam Kumaran: like, for… this is one thing that I’ll think about later, is, like.
133 00:14:52.210 ⇒ 00:15:11.009 Uttam Kumaran: it needs to feel very painful for us as a company to bring on new people. Like, yes, there’s a joy, because I think some of the people we’re bringing on now, there’s actually, like, a huge hole, but over time, every new person should be a fight to, like, bring them in. Like, it should really… we should be questioning, like.
134 00:15:11.720 ⇒ 00:15:16.030 Uttam Kumaran: Can we get extra time from anyone that exists? Is this something with AI we can do?
135 00:15:16.180 ⇒ 00:15:18.310 Uttam Kumaran: Hiring should not be the first
136 00:15:18.620 ⇒ 00:15:37.039 Uttam Kumaran: go to. I know there’s great people out there, and we could totally get them, and they’re gonna love working here, but it will… it’s gonna go to affect, like, the money that’s available, and it’s an extra node in the network of information that has to go through, right? That’s how I think about things as well.
137 00:15:37.680 ⇒ 00:15:46.969 Henry Zhao: And another thing I was thinking is, like, the clients that we go for, right, like, there’s a lot of startups, there’s a lot of smaller companies that they need data analytics, they need kind of what we do.
138 00:15:47.070 ⇒ 00:15:55.519 Henry Zhao: But if they want to hire even someone junior, right, that’s gonna be, like, $100K out of their budget. Whereas, we should be filling in those gaps for a lot of different companies.
139 00:15:55.520 ⇒ 00:15:56.000 Uttam Kumaran: Yes.
140 00:15:56.000 ⇒ 00:16:01.929 Henry Zhao: They can pay us less than that, and we can ramp up really quickly, we can be flexible, right? Like, I think.
141 00:16:01.930 ⇒ 00:16:07.800 Uttam Kumaran: Yeah, like, I would actually like to go to market with something that’s a little bit cheaper than we are now. The problem is…
142 00:16:08.850 ⇒ 00:16:15.600 Uttam Kumaran: Is that we… we have to find these… we have to create these really well-documented playbooks.
143 00:16:15.600 ⇒ 00:16:18.660 Henry Zhao: And we have to be very AI efficient for me to open.
144 00:16:18.960 ⇒ 00:16:21.740 Uttam Kumaran: The door to lowering our costs, because
145 00:16:22.070 ⇒ 00:16:39.030 Uttam Kumaran: it’s really expensive to run this business, and to just achieve, like, any sort of margin. To any of those clients, one, often don’t pay on time, and are very cheap. I actually want to go work with them, because my whole career is in startups. I want to go have an offering for them.
146 00:16:39.080 ⇒ 00:16:43.919 Uttam Kumaran: But it… we need to be very crisp on what we do, what we don’t do, and…
147 00:16:44.400 ⇒ 00:16:53.439 Uttam Kumaran: We have to price towards, like, an outcome more strategically than we would do for these big, bigger clients, just because of the risk, right?
148 00:16:53.440 ⇒ 00:16:57.689 Henry Zhao: Yeah, I feel like you could charge the same price, because it’ll be cheaper than them hiring somebody.
149 00:16:57.870 ⇒ 00:17:02.339 Uttam Kumaran: I mean, that’s what I say, but dude, where it comes down to is they’re bad clients.
150 00:17:02.530 ⇒ 00:17:10.269 Uttam Kumaran: They don’t pay on time, and they’re all over the place. And commonly, they churn because they fail. We’ve had that happen.
151 00:17:11.040 ⇒ 00:17:13.670 Uttam Kumaran: So for that reason, I sort of, like…
152 00:17:14.170 ⇒ 00:17:21.339 Uttam Kumaran: we, we across the board don’t work with startups. Like, all the startups we work for right now, like, default, Caitlin is, like, a friend of mine.
153 00:17:21.890 ⇒ 00:17:22.280 Henry Zhao: Okay.
154 00:17:22.280 ⇒ 00:17:24.880 Uttam Kumaran: And they raised, like, $20 million.
155 00:17:25.200 ⇒ 00:17:26.260 Henry Zhao: So, like.
156 00:17:26.260 ⇒ 00:17:28.720 Uttam Kumaran: It checks, like, 2 of my boxes.
157 00:17:28.720 ⇒ 00:17:29.170 Henry Zhao: Okay.
158 00:17:29.170 ⇒ 00:17:37.379 Uttam Kumaran: She’s like, I need to know someone there, and they need to, like, actually be in the money. We were working with a lot of startups, we struggled because
159 00:17:37.640 ⇒ 00:17:51.809 Uttam Kumaran: they don’t, like, they don’t have time to meet, like, they don’t give feedback on time, sometimes it’s, like, up and down, up and down, like, our business can’t be… like, we’re not a mirror reflection of the companies we’re working for. We need to be, like, the rock.
160 00:17:51.930 ⇒ 00:17:55.900 Uttam Kumaran: Like, disable, you know, but sometimes, like.
161 00:17:56.670 ⇒ 00:18:06.129 Uttam Kumaran: those types of clients are extremely hard. I think one is we… I can work with them, I think you can work with them, I think Robert can, but can our… can the rest of our team? Not sure.
162 00:18:06.150 ⇒ 00:18:18.350 Uttam Kumaran: Right, so one thing is, like, maybe we need to build more of an arm of, like, how to work with those startups, like, how to lean more communication, less delivery, or whatever. And, like, how do we actually…
163 00:18:18.470 ⇒ 00:18:25.610 Uttam Kumaran: Like, or maybe it’s like, hey, given those risks, maybe we should price even higher. And the ones that go for it, then at least we make
164 00:18:25.830 ⇒ 00:18:31.890 Uttam Kumaran: Or we have stricter terms, like, for starters, for example, I now do net 15, and I try to get them to pay up front.
165 00:18:32.360 ⇒ 00:18:37.170 Uttam Kumaran: Because I have to mitigate against the fact that, like, dude, people wouldn’t pay us for 90 days.
166 00:18:37.650 ⇒ 00:18:46.249 Uttam Kumaran: I don’t… like, there’s not, like, money, like, sitting to, like, go 90 days without, like… for work we did so long, it was horrible, and…
167 00:18:46.250 ⇒ 00:18:48.459 Henry Zhao: It was something that I had to just, like…
168 00:18:48.630 ⇒ 00:19:01.349 Uttam Kumaran: Nicks for a bit, but they’re also, again, like, there is a class, I think you, me, Robert, like, Demolade is like that, Awash, where we can just sort of walk into any situation, figure it out. So maybe there is, like, a…
169 00:19:01.690 ⇒ 00:19:17.729 Uttam Kumaran: a, like, Navy SEAL group that I can go be like, hey, if you want… if you guys look like you need the Navy SEAL group, that’s gonna be $500 an hour, and if they go for it, they go for it. And, like, that’s sort of how we mitigate against that, because I also think there is something beneficial to being, like, a special situations group.
170 00:19:17.840 ⇒ 00:19:21.439 Uttam Kumaran: Right, where it’s whatever, like, there’s not a lot of consultancies
171 00:19:21.590 ⇒ 00:19:28.840 Uttam Kumaran: consistency will take that business and then fail. And I think our business now, we may take that and fail, but if we have a group tailored toward, like.
172 00:19:29.080 ⇒ 00:19:40.680 Uttam Kumaran: working with chaotic startups, and we, like, all get a sense of, like, what that means, and how to deal with it, and it’s a… we charge an appropriate amount relative to risk, I’m game to do that.
173 00:19:40.740 ⇒ 00:19:59.259 Uttam Kumaran: And, like, that is not… I don’t think companies like that exist. Like, there are people that say we only work with startups. It’s usually because that’s the only people they can sell to. Like, because nobody else will pick up their phone. For us now, we’re getting… people are picking up our phone that are way, way bigger. I still would like to service those startups, because, like.
174 00:19:59.460 ⇒ 00:20:05.020 Uttam Kumaran: we could easily sell to startups. I think they’re just hard to execute, and, like.
175 00:20:05.510 ⇒ 00:20:07.129 Uttam Kumaran: To do what they need, you know?
176 00:20:07.710 ⇒ 00:20:14.100 Henry Zhao: Yeah. Yeah. I just think when we have a strong, like, rockstar group, like, a team, we’ll be able to take on more clients and more work, so…
177 00:20:14.100 ⇒ 00:20:14.740 Uttam Kumaran: Yeah.
178 00:20:14.930 ⇒ 00:20:15.250 Henry Zhao: Yeah.
179 00:20:15.460 ⇒ 00:20:21.589 Uttam Kumaran: I agree. I mean, I want us to go take on… for me, the next big milestone is, like, I want to sell more Eden’s, like.
180 00:20:22.030 ⇒ 00:20:30.300 Uttam Kumaran: 50K a month, where we’re running the whole show, I want to also go take on more difficult data work, like…
181 00:20:30.650 ⇒ 00:20:45.429 Uttam Kumaran: On the AI side, this could be fine-tuning and training models. On the data side, this could be real data science, machine learning, stuff that takes longer, requires, like, more skilled labor, like, really gets us to expand what we’re doing.
182 00:20:47.020 ⇒ 00:20:54.480 Uttam Kumaran: I would say on the AI side is where the demand is really, really high. So then what you’ll kind of see as I start to loop you into some stuff.
183 00:20:54.630 ⇒ 00:21:01.369 Uttam Kumaran: you’ll see that, like, it’s actually pretty easy to get started with some of these AI prototypes. It’s hard to get it, like, right every time.
184 00:21:01.440 ⇒ 00:21:06.239 Henry Zhao: But our pitch and the amount of work we’ve done over the last year is we’re… we’re, like.
185 00:21:06.490 ⇒ 00:21:19.839 Uttam Kumaran: we’re pretty far compared to other agencies that are also doing AI. Also, for the AI, a lot of it is data work. A lot of it is data engineering work. A lot of it is measuring how the agent is working, or the AI systems are working. It’s all just, like.
186 00:21:19.940 ⇒ 00:21:36.320 Uttam Kumaran: normal data stuff, so it, like, rhymes with what we do really well. So it’s not like we’re going to something that’s so far away. And for the most part, I think I want you to see how we could pitch AI solutions to our existing data clients.
187 00:21:36.420 ⇒ 00:21:44.139 Uttam Kumaran: And similarly, when we go into AI clients, how can we pitch them on data stuff? I guarantee you there’s all this BS in Eden that we could go automate.
188 00:21:44.260 ⇒ 00:21:50.390 Uttam Kumaran: But it takes some people that are sitting in both camps to kind of, like, see that and propose those opportunities.
189 00:21:50.560 ⇒ 00:22:01.889 Uttam Kumaran: So that’s what I’m kind of excited to see. Like, Mustafa and Casey are starting to do some data work, too. I’m excited, like I said today, for some people on the data side to get interested, and, like.
190 00:22:02.030 ⇒ 00:22:06.050 Uttam Kumaran: See how we can use it in our company to speed up our own data work, but then also
191 00:22:06.390 ⇒ 00:22:20.169 Uttam Kumaran: start building AI stuff, because it’s… it’s actually, like, such a new field that I don’t think you’re… as long as you start at some point, you’re not going to be very behind, like, where most of the crowd is, you know? We’ve only been do… I’ve only been doing this
192 00:22:20.250 ⇒ 00:22:29.270 Uttam Kumaran: I’ve been using ChatGPT for, like, 2 years, and then the only reason I’m ahead is because I’ve been using it to build a company, and I just, like, search for the next thing. But 2 years is, like.
193 00:22:29.610 ⇒ 00:22:38.349 Uttam Kumaran: Two years… this is, like, 10% of my, like, brain, so we’re not that… we’re not as far as you think. I think you’ll find it pretty okay.
194 00:22:38.730 ⇒ 00:22:40.239 Henry Zhao: Okay, yeah, sounds good to me.
195 00:22:40.240 ⇒ 00:22:41.829 Uttam Kumaran: I’d love to be part of that.
196 00:22:42.190 ⇒ 00:22:44.649 Uttam Kumaran: Okay, how else can I… can I help?
197 00:22:45.160 ⇒ 00:22:55.080 Henry Zhao: Yeah, so about the, like, fixed versus whatever, I just remember my contract ends, I think, the end of this month. Cool. So, I don’t know if you need to get in a call with me and Robert to just…
198 00:22:55.180 ⇒ 00:22:58.569 Henry Zhao: Figure out where we go from there, or just don’t want to know how that…
199 00:22:59.130 ⇒ 00:23:02.840 Uttam Kumaran: Yeah, let me… let me talk to him. I mean, if, like.
200 00:23:03.110 ⇒ 00:23:10.960 Uttam Kumaran: I think there’s a couple options. One is, like, we maintain, sort of, the current pricing. I would say if we move to something fixed.
201 00:23:11.090 ⇒ 00:23:23.879 Uttam Kumaran: like, I would see whether you’re flexible at all on rate, given that we can, like, make a longer commitment. But also, like, dude, I really enjoy working with you. I want to make sure that you’re committed, and this is a place that you want to be at.
202 00:23:24.100 ⇒ 00:23:28.849 Uttam Kumaran: And that we can start to also pay you more over the long term.
203 00:23:29.000 ⇒ 00:23:40.160 Uttam Kumaran: So that’s kind of, like, where my head’s at. Net-net, like, we can keep it the same. If you think, like, hey, I want some more commitment, and I want to know that, like, this amount is going to come in, we can move to something more fixed.
204 00:23:41.660 ⇒ 00:23:57.000 Uttam Kumaran: And then, yeah, I think over… what you’re gonna see over this quarter is that… and next quarter is we’re gonna move to do, like, sort of structured rate increases based on performance, both feedback and… and… and performance. So, performance meaning just, like.
205 00:23:57.060 ⇒ 00:24:12.330 Uttam Kumaran: Are we getting… are we getting work done? It’s… that… that is more to find folks that aren’t getting stuff done, really. It’s like, are we… are we taking a reasonable amount of tickets on, reasonable amount of points, are we getting them done? Second is going to be feedback from clients and the team.
206 00:24:13.380 ⇒ 00:24:19.639 Uttam Kumaran: And then all of that is gonna go to benefit the most effective people. We’re also gonna start, hopefully, doing, like, more…
207 00:24:19.790 ⇒ 00:24:29.189 Uttam Kumaran: like, bonuses based on, like, company metrics. So… and I don’t know whether we’re gonna do this on a team-by-team or just company, but, like, if we’re able to hit
208 00:24:29.280 ⇒ 00:24:39.939 Uttam Kumaran: profit goals, or if we’re able to hit, sort of different milestones, I want to reward the best people. And I think over time, this is going to be a company where the best people
209 00:24:39.960 ⇒ 00:24:50.219 Uttam Kumaran: you’re gonna hopefully make more than you will in the other market, and the people that aren’t so good are gonna make less, and naturally churn out. And that’s gonna be the cycle.
210 00:24:50.280 ⇒ 00:24:55.570 Uttam Kumaran: I just, like, word vomited a lot, like, that’s kind of, like, where… what I’m thinking long-term.
211 00:24:56.350 ⇒ 00:25:04.440 Henry Zhao: Yeah, makes sense. So yeah, if you can talk to Robert, I’ll talk to Robert as well, and just let me know what you guys are thinking, we can figure something out before the end of the month.
212 00:25:04.750 ⇒ 00:25:05.980 Uttam Kumaran: Okay, okay, perfect.
213 00:25:07.180 ⇒ 00:25:08.140 Uttam Kumaran: Alright, man.
214 00:25:08.310 ⇒ 00:25:09.520 Uttam Kumaran: That’s all I had.
215 00:25:09.520 ⇒ 00:25:11.960 Henry Zhao: That’s it on my end as well. Thanks for the time.
216 00:25:12.290 ⇒ 00:25:13.670 Uttam Kumaran: Alright, thank you so much.
217 00:25:13.680 ⇒ 00:25:14.649 Henry Zhao: Thank you, bye-bye.