Meeting Title: Stitch Data Warehouse Project Check-in Date: 2026-01-08 Meeting participants: Samuel Roberts, Uttam Kumaran, Bobby Palmieri, zacfromson
WEBVTT
1 00:00:08.300 ⇒ 00:00:09.560 Uttam Kumaran: Ayo.
2 00:00:09.560 ⇒ 00:00:10.530 Samuel Roberts: a…
3 00:00:11.930 ⇒ 00:00:13.180 Uttam Kumaran: Feeling okay?
4 00:00:14.550 ⇒ 00:00:17.650 Samuel Roberts: I can’t get warm, but that’s, like, the only symptom I have right now.
5 00:00:17.950 ⇒ 00:00:18.640 Uttam Kumaran: Okay.
6 00:00:18.640 ⇒ 00:00:25.399 Samuel Roberts: So it’s just like, I’m just cold, and I don’t like that. It’s a very ominous, like, I don’t know what’s gonna happen tomorrow kind of thing.
7 00:00:25.400 ⇒ 00:00:29.910 Uttam Kumaran: I heard there’s a lot of flu right now, actually. Like, I heard there’s, like, some type of major flu.
8 00:00:29.910 ⇒ 00:00:37.469 Samuel Roberts: That’s what I… I was talking to my friend, and he’s like, did you get the new flu? And I was like, oh, no, I hope not. I’ve been talking about getting my shot for, like, 3 weeks, and I just haven’t.
9 00:00:37.470 ⇒ 00:00:40.830 Uttam Kumaran: I just see it on Twitter, and, like, I don’t know whether it’s, like.
10 00:00:41.070 ⇒ 00:00:44.830 Uttam Kumaran: AI, or fake, or, like, real. I’m like, okay.
11 00:00:45.600 ⇒ 00:00:50.990 Samuel Roberts: Yeah, normally it’s cold in the attic, so I was like, oh, it’s just… and it’s, like, 70 degrees in here, and I’m like, oh, that’s not good.
12 00:00:51.760 ⇒ 00:00:57.139 Samuel Roberts: But… Doing okay besides that, so hopefully, if that’s… that’s it for now, we’ll be alright.
13 00:00:57.720 ⇒ 00:00:58.819 Samuel Roberts: How you guys doing?
14 00:00:59.150 ⇒ 00:01:07.390 Bobby Palmieri: My… my big hack, Zach let me say this on, like, 6 calls, I saw it on Twitter as well. Gargle with green tea.
15 00:01:07.680 ⇒ 00:01:15.440 Bobby Palmieri: Supposedly, like, 3 times a day, it cuts your chance of contracting sickness by 60% if you think you’ve been exposed.
16 00:01:15.440 ⇒ 00:01:16.060 Samuel Roberts: Oh, wow.
17 00:01:16.950 ⇒ 00:01:20.499 Uttam Kumaran: I’m gonna take preemptive measure, then, later.
18 00:01:20.500 ⇒ 00:01:21.330 Samuel Roberts: Yes.
19 00:01:21.750 ⇒ 00:01:23.180 Bobby Palmieri: Agreed.
20 00:01:23.890 ⇒ 00:01:28.519 Uttam Kumaran: That’s funny. Yeah, I heard it’s really bad, right? It’s like all-season change, I feel like, so…
21 00:01:28.520 ⇒ 00:01:29.340 Samuel Roberts: Yeah.
22 00:01:29.970 ⇒ 00:01:36.369 Uttam Kumaran: Thankfully, it’s warm here in Texas still, like, so there’s no… there’s no weather change. I am… it’s the same weather that’s always been.
23 00:01:37.030 ⇒ 00:01:43.520 Uttam Kumaran: It’s… it’s just not hot, it’s just normal. So, yeah.
24 00:01:44.270 ⇒ 00:01:51.940 Uttam Kumaran: Okay, cool. Well, yeah, thanks everyone for joining. Yeah, I think today I really wanted to just chat a little bit about, sort of our
25 00:01:52.000 ⇒ 00:02:08.299 Uttam Kumaran: proposal for trying to speed up what parts, you know, we… we could. I think, just to give it, like, a recap of, like, four priorities right now. So, one is, I think we’re gonna have a lot to share, tomorrow of, like, core MCPs, admin UI, like.
26 00:02:09.090 ⇒ 00:02:11.009 Uttam Kumaran: A good amount of the…
27 00:02:11.380 ⇒ 00:02:31.169 Uttam Kumaran: V1 sort of ready to go. I don’t know, so that’s, like, kind of, like, we’re still gearing up to basically have most of that done by mid-next week. I think I’m gonna pull up, sort of, the Gantt on my side, and kind of share, like, where we’re going through next, and then really wanted to…
28 00:02:31.400 ⇒ 00:02:46.359 Uttam Kumaran: to just talk through, like, if we’re… if we felt like all the pieces, are necessary, what pieces, you know, we… we want to sort of move off and kind of talk through, like, what we could do to parallelize more. Yeah.
29 00:02:46.360 ⇒ 00:02:48.129 Bobby Palmieri: I think even before… before you.
30 00:02:48.130 ⇒ 00:02:48.460 Uttam Kumaran: Sure.
31 00:02:48.870 ⇒ 00:03:03.309 Bobby Palmieri: pull that up. I think, just as kind of, like, an overview to make sure that, like, things don’t get lost in Slack DMs or messages, like, yeah, I think we’re super happy with where we’re at. Cool. I think there’s…
32 00:03:03.680 ⇒ 00:03:08.229 Bobby Palmieri: urgency on our side to run fast. You know.
33 00:03:08.230 ⇒ 00:03:25.549 zacfromson: The sooner we can implement this with the team, the more effective we can get buy-in, we can start making… and it’s obviously not your guys’ fault, but we were trying to build this product for months ahead of time, so we’re already, I guess, structurally delayed, and the sooner we can get even just the… some of these key pieces into the team, like…
34 00:03:26.190 ⇒ 00:03:31.740 zacfromson: financially, we’d rather just compound it faster, if possible, than I think, like, Save money.
35 00:03:31.740 ⇒ 00:03:46.970 Bobby Palmieri: if we’re gonna spend 30 grand, 50 grand, right, like, I’d rather spend it all in January than, you know, 10K a month for 5 months. You know, in that regard, I think that we see the…
36 00:03:47.040 ⇒ 00:03:52.059 Bobby Palmieri: the future of what Stitch will be, and what it will do for the agency, and like…
37 00:03:52.670 ⇒ 00:03:59.840 Bobby Palmieri: we’re probably gonna build 100 different tools this year, right? Like, you guys will always have work.
38 00:03:59.910 ⇒ 00:04:13.010 Bobby Palmieri: with us, like, there’s just so much shit that we want to build, and I think the reason that I keep stressing the data warehouse is, like, all of this falls into place once the data warehouse is done. Yeah. And, you know, I think…
39 00:04:13.210 ⇒ 00:04:20.749 Bobby Palmieri: Yeah, so I think that’s, you know, kind of where we’re at. We just want to… want to run faster on, like, getting MCP set up, getting Nano Banana.
40 00:04:21.019 ⇒ 00:04:25.959 Bobby Palmieri: integrated with, like, what we have, because, like, I mean, I see what the team is…
41 00:04:26.390 ⇒ 00:04:37.509 Bobby Palmieri: is using it for, and, like, you know, just based on API credits, it’s, like, the most used tool we’ve ever had. So, like, we just want to get that, like, you know, into…
42 00:04:37.620 ⇒ 00:04:39.360 Bobby Palmieri: stitch, so that, like.
43 00:04:39.710 ⇒ 00:04:56.420 Bobby Palmieri: employees are logging into Stitch, they’re starting to use MCP, they’re using this in, like, a central place that has, like, a repository, and then we have a data warehouse where, like, hey, there’s a report that, like, we want to start using. I’ll show you one tool, Sam, which I think might just be helpful for,
44 00:04:56.780 ⇒ 00:05:00.340 Bobby Palmieri: For your context on Facebook authentication, I’m in,
45 00:05:00.340 ⇒ 00:05:00.690 Samuel Roberts: Yeah.
46 00:05:00.880 ⇒ 00:05:08.339 Bobby Palmieri: for, like, Facebook ad agency owners, and they had, like, built a tool for the community. It’s super simple, but, like.
47 00:05:09.360 ⇒ 00:05:25.979 Bobby Palmieri: it’s so simple, but it will be used across the agency, and I think, like, when we get to a point where we either have a data warehouse or have this stood up, like, these are the things that, like, I just want to be able to either have you guys rip in Cloud Code, or rip, like, doesn’t matter to me. I also, like…
48 00:05:26.720 ⇒ 00:05:47.179 Bobby Palmieri: I want you guys to know, like, as much as I’m like, hey, we want to be able to push this in, like, Claude or whatever, I know that there’s shit that, like, we will need the system architect behind, so it’s not like a, why aren’t you guys running faster? I can do this and reply. It’s more of just, like, there are some things that I do think could be done in Cloud, and, like, I think will be helpful for the agency.
49 00:05:47.360 ⇒ 00:05:53.380 Bobby Palmieri: So, all that to say, like, you know, I think we’ve been super happy with the partnership, especially coming out of…
50 00:05:53.440 ⇒ 00:06:08.670 Bobby Palmieri: our last relationship, and, like, we just want to push fast, and I see the data warehouse as, like, the biggest hurdle for you guys to run faster, for Zach and I to be able to push shit, and, like, if we need to…
51 00:06:08.720 ⇒ 00:06:20.950 Bobby Palmieri: pull up some, you know, spend and invoices for you guys, like, let’s do that so that you guys can start, you know, building out the data warehouse this week, or whenever.
52 00:06:20.950 ⇒ 00:06:29.370 Bobby Palmieri: But, like, I just see that as, like, the hurdle where, like, we’re climbing up this mountain, and as soon as that’s there, then it’s like, okay, we’re in the car, and we’re running.
53 00:06:29.370 ⇒ 00:06:29.890 Uttam Kumaran: Yeah.
54 00:06:29.890 ⇒ 00:06:35.220 Bobby Palmieri: you know, in that regard. I don’t know how you… how you feel, or, like, is that… is that accurate?
55 00:06:35.700 ⇒ 00:06:51.290 Uttam Kumaran: Yeah, I would say, Bill, so there’s kind of like, you know, when we walk into a situation, there’s always a couple things. Like, one, you guys had a previous partner, so I know there’s, like, already delays built in. Second, we just had holiday slowdowns, so that’s already built in. And third, it’s like.
56 00:06:51.290 ⇒ 00:07:04.529 Uttam Kumaran: anything in the backend I know is not visually apparent, it doesn’t seem like progress. So, as a product person, I know all those things are at play. For me, the biggest thing is, one, all of the speed that you guys need
57 00:07:04.530 ⇒ 00:07:14.079 Uttam Kumaran: down the line is going to be dependent on getting this first part right. And, like, I don’t… I know, like, I don’t… I’m not someone, and Sam can attest, who, like.
58 00:07:14.180 ⇒ 00:07:29.319 Uttam Kumaran: is, like, very careful. I’m kind of like, let’s move really, really fast. But there’s core parts of authentication, scaling, deployment, that if we mess up, we’re gonna be in the same spot as those… those people we’re in. So, like.
59 00:07:29.320 ⇒ 00:07:40.760 Uttam Kumaran: You know, so that’s one piece. Second is, like, I… I told Sam, like, I want Pranav, like, kind of, like, basically dedicated to, like, ripping stuff for y’all. I think you’ll see all the progress we made.
60 00:07:40.760 ⇒ 00:07:42.899 Uttam Kumaran: tomorrow. I also do…
61 00:07:42.900 ⇒ 00:07:54.899 Uttam Kumaran: like, want to loop in Casey on our team, to basically start to take on the data work. We basically went through and mapped out all the stuff on Phase 2, and that’s the bulk
62 00:07:54.920 ⇒ 00:08:05.690 Uttam Kumaran: of our time is the data infrastructure and the cohort… the kind of first model of the cohorting model. After that, the UI, the dashboard, the settings, like.
63 00:08:06.050 ⇒ 00:08:19.920 Uttam Kumaran: I don’t… I’m not worried about. And actually, a lot of those pieces, like, if you guys… if we have the endpoints and you guys vibe code a first version, you’ll get a 60% there, we take that and… and put that in. And so, like, I… yeah.
64 00:08:20.320 ⇒ 00:08:27.119 Bobby Palmieri: Yeah, the only thing, and Zach, I sent you a message on this, I don’t know if you responded or not, but,
65 00:08:27.380 ⇒ 00:08:28.900 Bobby Palmieri: there’s, like…
66 00:08:29.090 ⇒ 00:08:34.029 Bobby Palmieri: when we look at, kind of, the Gantt chart, there’s one, like… here, let me just show you, like.
67 00:08:34.030 ⇒ 00:08:41.170 Uttam Kumaran: Sure, yeah, yeah, yeah. And I think there’s a lot of pieces of the Gantt that I think you guys will be able to ship.
68 00:08:41.320 ⇒ 00:08:49.880 Uttam Kumaran: It’s… it is… it’s also… this is, like, a new way of working, also, so it’s… it’s tough for… for us to forecast versus…
69 00:08:50.450 ⇒ 00:09:04.749 Uttam Kumaran: give an answer, like, one… the worst thing I could do is be, like, yeah, we’re gonna get this all done in, like, 4 weeks, and then, like, it just doesn’t happen. And so that’s, like, that’s really, like, how… why I’m kind of biased a little bit to make sure we hit the target, you know?
70 00:09:04.750 ⇒ 00:09:12.310 Bobby Palmieri: Yeah, no, totally, and I appreciate that. I think the one thing that, when we start talking about Phase 2, like, obviously we need to get the data warehouse.
71 00:09:12.430 ⇒ 00:09:21.859 Bobby Palmieri: Yeah. And, and, I don’t know if this is gonna be the right one.
72 00:09:22.290 ⇒ 00:09:28.650 Bobby Palmieri: But one thing that, you know, we were kind of looking at is…
73 00:09:30.980 ⇒ 00:09:33.849 Bobby Palmieri: Like, this is what the team is currently using.
74 00:09:34.240 ⇒ 00:09:43.840 Bobby Palmieri: Yeah. Right? So, like, if we look at our tracker here, I don’t know if this is the right version, but, like, when we look at daily forecasting, right, like.
75 00:09:44.250 ⇒ 00:09:56.199 Bobby Palmieri: this is what we’ve built in a spreadsheet. Obviously, we want to have, like, the monthly forecast and, like, the scenarios, but I’m wondering if, like, just in the speed of what the team is using.
76 00:09:56.600 ⇒ 00:10:01.450 Bobby Palmieri: We build just the monthly tracker where we just input the things, like.
77 00:10:01.450 ⇒ 00:10:01.920 Uttam Kumaran: Okay.
78 00:10:01.920 ⇒ 00:10:06.540 Bobby Palmieri: now, and then, like, in the background, we can work on Python and rolling that out.
79 00:10:06.540 ⇒ 00:10:07.030 Uttam Kumaran: Okay.
80 00:10:07.030 ⇒ 00:10:17.819 Bobby Palmieri: But, like, just getting a daily tracker compared to forecast. Because this is what the team is using, and Zach, transparently, like, none of our clients
81 00:10:18.460 ⇒ 00:10:21.679 Bobby Palmieri: Or… we’re not giving the forecast.
82 00:10:22.280 ⇒ 00:10:26.449 Bobby Palmieri: Based on, like, our cohort analysis, like, we were being given
83 00:10:27.220 ⇒ 00:10:35.470 Bobby Palmieri: So, like, whatever our model says doesn’t really matter in the short term. I think long term, we want to go to them and be like, hey, based on cohort analysis, like.
84 00:10:35.470 ⇒ 00:10:36.390 Uttam Kumaran: Yes, okay.
85 00:10:36.390 ⇒ 00:10:36.790 Bobby Palmieri: track.
86 00:10:36.790 ⇒ 00:10:37.500 Uttam Kumaran: Period.
87 00:10:38.150 ⇒ 00:10:39.400 Bobby Palmieri: So, yeah, I think.
88 00:10:39.400 ⇒ 00:10:48.629 Uttam Kumaran: Yeah, I mean, look, this would… if we drive… so, in that spreadsheet, I assume it’s just, like, a kind of flatline, like, linear… or whatever the percentage input forecasts, right?
89 00:10:48.630 ⇒ 00:10:52.589 Bobby Palmieri: Yeah, so it’s just like, you know, spend is, you know, here, and then it’s 2%.
90 00:10:53.050 ⇒ 00:11:09.420 Uttam Kumaran: So, yeah, I mean, this is, like, the… this is, like, much simpler than the cohort sort of Python thing. Like, ultimately, there’s kind of… there’s two big pieces to this next phase. There’s one, like, it’s the data warehouse, the data ingestion. The cohort modeling is, like.
91 00:11:09.780 ⇒ 00:11:23.840 Uttam Kumaran: the actual, in addition to the data warehouse stuff, the biggest piece, because we have to set up, like, the Python script to get the cohort modeling right, and then pipe the results out. The UI is not what I’m worried about at all.
92 00:11:23.840 ⇒ 00:11:24.260 Bobby Palmieri: Yeah.
93 00:11:24.260 ⇒ 00:11:37.220 Uttam Kumaran: And like, I, you know, so the UI is not what I’m worried about. It’s actually just, like, getting that model right. So if you’re, like, you know, skip the model for now to get this live, then it probably will cut
94 00:11:37.370 ⇒ 00:11:43.819 Uttam Kumaran: like, a week and a half. And then in parallel, like… but again, getting this right doesn’t mean…
95 00:11:44.110 ⇒ 00:12:01.859 Uttam Kumaran: it’s, like, this in, Stitch with the input fields and, like, being able to, you know, hook this into our Slack reports and stuff like that. It’s basically, like, yeah, whatever version of this, and so we still need the data warehouse, it’s just we need a much simpler model to power.
96 00:12:02.120 ⇒ 00:12:06.359 Bobby Palmieri: Yeah, and I think… and I think, you know, we do want to get to the point where we are
97 00:12:06.540 ⇒ 00:12:13.910 Bobby Palmieri: doing this, but I think that, like, now, if I think about what we need in February versus, like, what we need
98 00:12:14.210 ⇒ 00:12:15.260 Bobby Palmieri: You know.
99 00:12:15.260 ⇒ 00:12:15.610 Uttam Kumaran: Okay.
100 00:12:16.310 ⇒ 00:12:20.739 Bobby Palmieri: Q2. Zach, what are your thoughts on that? We can just kind of chat through it live here.
101 00:12:22.600 ⇒ 00:12:33.000 zacfromson: Yeah, I mean, I… I think, you know, ones and zeros, that, like, having something functional working as an event makes sense, and we can layer more behind it as we go. I think, like, just getting…
102 00:12:33.840 ⇒ 00:12:44.719 zacfromson: some of this in here and standardize for us as an agency makes sense, and maybe it’s not as automated, and we can send, like, better versions as we go, I guess, Bobby, rather than trying.
103 00:12:44.720 ⇒ 00:12:48.810 Bobby Palmieri: Yeah, my thought… my thought is, like, you know, basically, like, the growth…
104 00:12:48.810 ⇒ 00:12:49.170 zacfromson: But…
105 00:12:49.170 ⇒ 00:12:58.359 Bobby Palmieri: They’ll do the, you know, the monthly forecast of revenue goal, spend goal, you know, for Meta, Google, etc. And then, you know, they can do the percentages by day.
106 00:12:58.660 ⇒ 00:13:02.780 Bobby Palmieri: And then it’s automated, you know, the actuals and variants.
107 00:13:03.310 ⇒ 00:13:03.950 Uttam Kumaran: Okay.
108 00:13:03.950 ⇒ 00:13:10.869 Bobby Palmieri: Versus, like, you know, the only difference that we’re saying is we’re taking out, like, these would be algorithmic of, like, hey, we’re expecting…
109 00:13:10.870 ⇒ 00:13:14.529 Uttam Kumaran: Yeah, so there’s, like, all the kind of scenarios…
110 00:13:14.530 ⇒ 00:13:14.980 zacfromson: Yeah.
111 00:13:14.980 ⇒ 00:13:16.850 Uttam Kumaran: There’s all the kind of scenario, like.
112 00:13:17.620 ⇒ 00:13:28.829 Uttam Kumaran: issue some stuff on pacing, like, building, like, sort of the cohort model. But for example, we still would need, sort of, the ability to do inputs, the ability to kind of hook this into reports,
113 00:13:28.900 ⇒ 00:13:45.620 Uttam Kumaran: have this, like, hooked up into the rest of the system, like, have settings, and then all the data warehouse stuff. So, we probably will cut, like, 30% of features if we just go with something simpler. The other thing is, like, as this is something where we will need, like.
114 00:13:45.840 ⇒ 00:14:03.379 Uttam Kumaran: dev staging, and production environments to test changes, right? The chat… the chatbot is something that we can just, like, keep iterating, and there’s… it’s, like, not… it’s, like, I would say, less fragile. Once you’re introducing Data Warehouse, and now there’s, like, 3 more steps before things get to the front end.
115 00:14:03.380 ⇒ 00:14:07.160 Bobby Palmieri: This is where, like, we’re gonna have an ability to push changes.
116 00:14:07.160 ⇒ 00:14:26.220 Uttam Kumaran: have y’all or a test group test, and then be like, okay, it’s going live to everybody. For me, Sam, like, a big win over the next, like, two weeks is, one, like, we bang out, like, the rest, we have, like, the prop library, Slack messages, stuff like that, and then we start getting people onboarded into that, plus
117 00:14:26.330 ⇒ 00:14:37.430 Uttam Kumaran: the nano Banana, because then what we’ll get out of that is that people are using the app. They have the chat, and they have the nano feature, and then we basically drive towards
118 00:14:38.160 ⇒ 00:14:39.300 Uttam Kumaran: this piece.
119 00:14:39.790 ⇒ 00:14:40.360 Bobby Palmieri: Yep.
120 00:14:40.470 ⇒ 00:14:44.040 Bobby Palmieri: Yeah, I think… I think that’s… Spot on.
121 00:14:44.550 ⇒ 00:14:48.649 Uttam Kumaran: And then, like, we… over time, I assume, like, we’re gonna move past
122 00:14:48.990 ⇒ 00:14:53.950 Uttam Kumaran: Noon, and just, like, do you guys have a sense of, like, Are we gonna go…
123 00:14:54.190 ⇒ 00:15:00.850 Uttam Kumaran: Is there, like… are we gonna go brand… is there a sequence of brands? Is it more, like, going after certain brand managers, and then, like.
124 00:15:01.170 ⇒ 00:15:02.290 Bobby Palmieri: No, I think…
125 00:15:02.290 ⇒ 00:15:03.519 Uttam Kumaran: As much as we can.
126 00:15:03.520 ⇒ 00:15:08.560 Bobby Palmieri: I think we’re gonna rip the band-aid off and go with 70 brands from day one.
127 00:15:08.770 ⇒ 00:15:12.080 Uttam Kumaran: Okay, so then we… it would be… we could get… if we could get that…
128 00:15:12.310 ⇒ 00:15:15.600 Uttam Kumaran: And then we’ll just need to go through key management and everything for that.
129 00:15:15.650 ⇒ 00:15:19.330 Samuel Roberts: And then if we… if you have a list of, like, who’s tapped in, who is…
130 00:15:19.400 ⇒ 00:15:24.210 Uttam Kumaran: Managing what brand. We just get that in a spreadsheet or something.
131 00:15:24.450 ⇒ 00:15:29.070 Uttam Kumaran: Yeah, we already have… And I assume we’ll be working with them for the keys and stuff, right?
132 00:15:29.270 ⇒ 00:15:33.509 zacfromson: Well, I’ll probably, like, just rip all the Klaviyo keys myself, basically.
133 00:15:33.510 ⇒ 00:15:33.990 Uttam Kumaran: Okay.
134 00:15:33.990 ⇒ 00:15:36.760 zacfromson: Just, like, from an access perspective, and there’s this.
135 00:15:36.760 ⇒ 00:15:37.280 Uttam Kumaran: Okay.
136 00:15:37.280 ⇒ 00:15:40.430 zacfromson: perspective, because I can just keep them hidden.
137 00:15:40.710 ⇒ 00:15:44.990 Samuel Roberts: Facebook and Google will be authenticated through a user, right? Right, yeah.
138 00:15:44.990 ⇒ 00:15:45.379 Uttam Kumaran: Yeah, yeah.
139 00:15:45.380 ⇒ 00:15:50.920 zacfromson: And then Shopify will probably just have Todd rip them all, Bobby, or have Anissa do it.
140 00:15:51.970 ⇒ 00:15:52.450 Bobby Palmieri: Right.
141 00:15:52.450 ⇒ 00:15:56.990 zacfromson: We need individual codes for… are we doing individual codes for Anthropic?
142 00:15:56.990 ⇒ 00:15:58.640 Bobby Palmieri: No, we’re not, or…
143 00:15:59.200 ⇒ 00:16:00.120 Uttam Kumaran: Oh, I don’t know.
144 00:16:00.530 ⇒ 00:16:04.450 Samuel Roberts: Now it’s set… per brand,
145 00:16:04.800 ⇒ 00:16:10.889 Samuel Roberts: Just because I want to match how you guys had it, but that’s flexible, too. Like, I can default it to a main one.
146 00:16:11.170 ⇒ 00:16:13.470 zacfromson: No, that’s fine, I mean, you just have a knee surgery.
147 00:16:13.690 ⇒ 00:16:15.810 zacfromson: Create one for a brand, Bobby?
148 00:16:15.810 ⇒ 00:16:16.390 Bobby Palmieri: Yep.
149 00:16:16.570 ⇒ 00:16:20.709 Bobby Palmieri: I think it’d be nice to see, kind of, usage across the board.
150 00:16:20.710 ⇒ 00:16:25.980 Samuel Roberts: Yeah, it’s the easiest way to do it until we start really layering in, like, more stuff into the ALMs.
151 00:16:25.980 ⇒ 00:16:30.730 Bobby Palmieri: Yeah. Okay. Yeah, that’s super easy. We have someone on the team, we have just a VA who can do it.
152 00:16:31.070 ⇒ 00:16:31.590 Uttam Kumaran: F.
153 00:16:32.180 ⇒ 00:16:36.739 Uttam Kumaran: So, let me go ahead and, like, remove a couple things. I mean, the other piece is, like.
154 00:16:37.160 ⇒ 00:16:43.980 Uttam Kumaran: as you guys use it, new features are gonna come up, and so that’s really what’s gonna add to the timeline. And so that’s just the one thing is, like.
155 00:16:44.510 ⇒ 00:16:52.919 Uttam Kumaran: we’re gonna… if… we’re just gonna have to have a conversation if there’s, like, new asks, to be like, do we want to do that now or later? And so that’s… that’s what we can talk about in our weeklies.
156 00:16:53.090 ⇒ 00:17:01.560 Uttam Kumaran: But overall, for me, it’s just, like, making sure that we can… if we can afford to have Pranav for a little bit more time, and we can get Casey, then I think we could…
157 00:17:02.160 ⇒ 00:17:14.370 Uttam Kumaran: we could… we could accelerate this. I feel like if we drive towards the dashboard, the dashboard that you shared, could probably end up removing, like, 30% of the features that we planned. Sam?
158 00:17:14.569 ⇒ 00:17:19.339 Uttam Kumaran: And we can put that scenario together. I also think most of Phase 3
159 00:17:19.790 ⇒ 00:17:23.469 Uttam Kumaran: Like, you could probably claud code.
160 00:17:24.270 ⇒ 00:17:26.969 Uttam Kumaran: I mean, I don’t know.
161 00:17:26.970 ⇒ 00:17:29.500 Bobby Palmieri: I guess at this point…
162 00:17:30.990 ⇒ 00:17:33.280 Uttam Kumaran: Yeah, yeah, go ahead.
163 00:17:33.310 ⇒ 00:17:37.010 Bobby Palmieri: I would probably… I’m not looking at this as, like.
164 00:17:37.200 ⇒ 00:17:42.949 Bobby Palmieri: Phase 1, 2, and 3 anymore. I guess it’s like, here are the features, I want to accomplish.
165 00:17:42.950 ⇒ 00:17:43.450 Uttam Kumaran: How much?
166 00:17:43.790 ⇒ 00:17:52.509 Bobby Palmieri: features as we can each week, right? Okay. What are we making progress towards? Like, you know, like, if I…
167 00:17:52.740 ⇒ 00:17:53.580 Bobby Palmieri: you know.
168 00:17:54.270 ⇒ 00:18:08.540 Uttam Kumaran: So, like, for example, Phase 1, all of those features are, like, just the platform in general, like, the ability to add keys, the ability for us to host on Railway, have the MCPs, like, that affects, like, basically everything we’re gonna do, you know?
169 00:18:08.660 ⇒ 00:18:09.340 Bobby Palmieri: Yup.
170 00:18:09.340 ⇒ 00:18:13.839 Uttam Kumaran: Exactly. But you’re right, the next set, our basket of, like, our basket of features.
171 00:18:14.210 ⇒ 00:18:21.940 Uttam Kumaran: And we kind of… we kind of… the way we’re gonna do things is, like, POC, MVP, and then you can just keep… we’ll just keep layering things on.
172 00:18:23.420 ⇒ 00:18:29.139 Uttam Kumaran: And for me, it’s just sort of, like, having a basket of hours to just push the train forward, you know?
173 00:18:29.140 ⇒ 00:18:36.579 Bobby Palmieri: Exactly, and I guess, like, you know, when we… when we had started, we wanted to get, like, MCP and Slack reports up, right?
174 00:18:36.580 ⇒ 00:18:37.020 Uttam Kumaran: Yeah.
175 00:18:37.020 ⇒ 00:18:53.779 Bobby Palmieri: And we wanted to get, like, forecasting, and then we wanted to get, like, the storyboards. Like, I think just as we’ve spent more time with it, spent more time with you guys, like, I don’t think we’re necessarily thinking about it in phases anymore. Okay. I think it’s like, hey, this is the operating system dashboard that we’re…
176 00:18:53.940 ⇒ 00:18:55.380 Bobby Palmieri: striving towards?
177 00:18:56.440 ⇒ 00:18:57.200 Bobby Palmieri: what, like.
178 00:18:57.200 ⇒ 00:18:57.650 Uttam Kumaran: Dad.
179 00:18:57.650 ⇒ 00:19:02.500 Bobby Palmieri: What do we want to build on? How much? How far are we? Okay. Yep, so it’s like, okay, so…
180 00:19:02.940 ⇒ 00:19:07.030 Bobby Palmieri: You know, we will have chat done, right?
181 00:19:07.810 ⇒ 00:19:26.529 Bobby Palmieri: And then we’ll have Slack reports done, and then it’s like, okay, what are we tackling next for sure? So it’s like, okay, we are gonna get the nano Banana tool set up as next, and then it’s like, okay, I think actually the most important thing that we do next is this media overview.
182 00:19:27.200 ⇒ 00:19:27.760 Uttam Kumaran: Yeah.
183 00:19:27.760 ⇒ 00:19:31.099 Bobby Palmieri: It’s like, okay, the meta dashboard,
184 00:19:31.310 ⇒ 00:19:33.939 Bobby Palmieri: You know, so on and so forth, and we just kind of have
185 00:19:34.700 ⇒ 00:19:36.830 Bobby Palmieri: This prototype, and we kind of…
186 00:19:37.110 ⇒ 00:19:42.139 Bobby Palmieri: Prioritize, like, what are the features that we want to add to this, and you guys…
187 00:19:42.390 ⇒ 00:19:44.970 Bobby Palmieri: You know, tell us what can get done in a week, a month, a year.
188 00:19:46.080 ⇒ 00:19:57.919 Uttam Kumaran: Yeah, I mean, the other… the only other thing I’ll say is, like, as we make ship products, we have to maintain them. So, like, there’s gonna be changes that are requested for all of these. So, like, again, I think if we’re just…
189 00:19:58.090 ⇒ 00:20:08.739 Uttam Kumaran: we’re just gonna be honest with, like, okay, we have, like, these 10 things we’re working on, there’s 3 more asks that came in, which… which to take? But I see… we’re parallelizing as much as possible, like, we’re…
190 00:20:08.740 ⇒ 00:20:09.410 Bobby Palmieri: Nope.
191 00:20:09.410 ⇒ 00:20:13.720 Uttam Kumaran: I guess Phase 2, Phase 3 is less about, like, this first, this next. It’s like…
192 00:20:14.490 ⇒ 00:20:18.659 Uttam Kumaran: as much as we can do, we’re gonna do.
193 00:20:18.780 ⇒ 00:20:32.799 Uttam Kumaran: And so, yeah, in terms of budget, I think we just need enough to bring on Pranav and bring on KC, Sam, and I think that’ll just basically cut, like… we’ll probably cut our timeline in, like, half.
194 00:20:33.250 ⇒ 00:20:34.750 Uttam Kumaran: Like, You know?
195 00:20:35.130 ⇒ 00:20:35.910 Bobby Palmieri: Great.
196 00:20:36.500 ⇒ 00:20:37.060 Uttam Kumaran: Yeah.
197 00:20:37.330 ⇒ 00:20:39.209 Uttam Kumaran: You know, and, and, yeah.
198 00:20:39.510 ⇒ 00:20:48.070 Bobby Palmieri: Yeah, no, I think… I think that sounds good. I’ll… I’ll just follow up in our DMs, just with, like, you know, what that looks like, and, like, just…
199 00:20:48.320 ⇒ 00:21:07.510 Uttam Kumaran: And I want you guys to know, like, you can, like, I want to… I do want to, like, get through this first hump, but again, like, you can scale us up and down. We just always operate in this sort of, like, pod model, so it’s sort of like, as you need, we can kind of move up and down. So if it’s, like, this bang as much out, and then you’re like, okay, we’re gonna…
200 00:21:08.150 ⇒ 00:21:12.159 Uttam Kumaran: slow down or ship, we can do that, so… I think that’s…
201 00:21:12.160 ⇒ 00:21:27.199 Bobby Palmieri: where we’re at is like, hey, let’s run at 10? Does it go down to 15? Like, let’s figure out where we’re at. But, like, you know, we just need to…
202 00:21:27.640 ⇒ 00:21:35.759 Bobby Palmieri: This is an investment for us that we know is critically important, but it’s critically important that we get it done quickly.
203 00:21:35.760 ⇒ 00:21:36.400 Uttam Kumaran: Yes.
204 00:21:36.710 ⇒ 00:21:37.839 Bobby Palmieri: In that regard.
205 00:21:38.210 ⇒ 00:21:39.120 Uttam Kumaran: Okay, okay.
206 00:21:39.120 ⇒ 00:21:52.439 zacfromson: Yeah, and I think once we get a couple of these things up, like, we can… the… I don’t want to say the pressure will be off to some extent, but I think, like, we’ll… Yeah. We can kind of, like, build out the feature flows, like, some of this stuff we’re just… even as just an operational agency, we’re just desperate for, and like…
207 00:21:52.440 ⇒ 00:21:53.259 Uttam Kumaran: Yeah, yeah.
208 00:21:53.260 ⇒ 00:21:58.970 zacfromson: We have, like, some future asks that we haven’t built out yet. We’ll be excited for them, but probably, you know, can have…
209 00:21:59.330 ⇒ 00:22:02.109 zacfromson: still a heavy pace, but maybe a little bit more moderate. I think just, like, we.
210 00:22:02.110 ⇒ 00:22:02.510 Uttam Kumaran: Yeah.
211 00:22:02.510 ⇒ 00:22:06.989 zacfromson: forecasting, we really need these LLMs, like, some of these things are just… Our team is, like.
212 00:22:07.290 ⇒ 00:22:08.470 zacfromson: Killing us for it.
213 00:22:08.840 ⇒ 00:22:19.310 Uttam Kumaran: Okay, so one thing I think, Sam, we heard is, like, one, we close out the rest of the phase one. I think we loop in Casey and have him port the Nano Banana.
214 00:22:19.620 ⇒ 00:22:21.579 Uttam Kumaran: Like, as fast as we can.
215 00:22:21.770 ⇒ 00:22:28.530 Uttam Kumaran: And then… between Casey and Pranav, whoever can start on the warehouse stuff starts on the warehouse stuff.
216 00:22:28.530 ⇒ 00:22:29.230 Samuel Roberts: Yampi.
217 00:22:29.230 ⇒ 00:22:31.150 Uttam Kumaran: And then we just go with that squad.
218 00:22:32.180 ⇒ 00:22:32.710 Samuel Roberts: Yeah.
219 00:22:33.300 ⇒ 00:22:49.439 Samuel Roberts: I think that sounds good. I mean, yeah, some of the other stuff, like the backend stuff, I’m already thinking about the different environments, because I want to make sure we’re getting it all set up, because the MCP’s talking everything. There’s a lot of wires to plug around, but I think once that’s kind of settled next week.
220 00:22:49.730 ⇒ 00:22:58.889 Samuel Roberts: by next week, hopefully settled today or tomorrow, if I can get it all stood up, then they’ll be able to just run and add things. It’s a little less, like, plumbing that I have to get done.
221 00:22:59.300 ⇒ 00:23:07.579 Uttam Kumaran: And then as soon as Phase 1’s done, I’ll have… we’ll ask Pranav to do a cloud code walkthrough, and then basically get…
222 00:23:08.190 ⇒ 00:23:10.600 Uttam Kumaran: Zach and Bobby, hopefully in the next, like, week.
223 00:23:10.980 ⇒ 00:23:13.949 Uttam Kumaran: Being able to just, like, build right on top of our stuff.
224 00:23:13.950 ⇒ 00:23:14.720 Bobby Palmieri: Cool.
225 00:23:14.920 ⇒ 00:23:18.579 Bobby Palmieri: The, sam, I think this might be helpful. I don’t know if it’s helpful.
226 00:23:18.580 ⇒ 00:23:19.130 Samuel Roberts: Yeah.
227 00:23:19.130 ⇒ 00:23:28.619 Bobby Palmieri: We were chatting about it. So this is, Foxwell Digital is just a group that I’m in, but they had, like, built this. So I just want to show you, like, how they are doing it.
228 00:23:28.920 ⇒ 00:23:33.939 Bobby Palmieri: So, like, basically, I connect… To add insights, right?
229 00:23:37.960 ⇒ 00:23:44.980 Bobby Palmieri: And then it just gives me a drop-down of, like, all the accounts that we have access to, and then this just has me…
230 00:23:45.270 ⇒ 00:23:49.779 Bobby Palmieri: Pull… you know, the analytics each time. They don’t store…
231 00:23:50.360 ⇒ 00:23:52.280 Bobby Palmieri: Anything,
232 00:23:53.270 ⇒ 00:23:58.960 Bobby Palmieri: And then it just shows, like… so for instance, like, this is something that, you know, we would probably want to…
233 00:23:59.130 ⇒ 00:24:06.750 Bobby Palmieri: ship as well, like, reach composition analysis, cost per reach, you know, it’s like a clean dashboard,
234 00:24:06.890 ⇒ 00:24:07.630 Uttam Kumaran: Okay.
235 00:24:07.730 ⇒ 00:24:12.920 Bobby Palmieri: But that’s how they’re doing… like, authentication.
236 00:24:13.540 ⇒ 00:24:14.300 Samuel Roberts: That makes sense.
237 00:24:14.300 ⇒ 00:24:24.849 Bobby Palmieri: So I guess the biggest thing would be, like, Zach and I can authenticate in, and then, like, in the animal stitch thing.
238 00:24:24.850 ⇒ 00:24:25.260 Uttam Kumaran: Yeah.
239 00:24:25.300 ⇒ 00:24:28.050 Bobby Palmieri: for that, which I think you… you had done.
240 00:24:28.920 ⇒ 00:24:29.480 Bobby Palmieri: Mine?
241 00:24:29.480 ⇒ 00:24:30.280 Samuel Roberts: Yeah.
242 00:24:30.280 ⇒ 00:24:31.320 Bobby Palmieri: So the…
243 00:24:31.320 ⇒ 00:24:32.200 Samuel Roberts: Clark?
244 00:24:32.340 ⇒ 00:24:34.789 Bobby Palmieri: No, I was gonna say, my question for you is.
245 00:24:35.510 ⇒ 00:24:37.899 Bobby Palmieri: this is an API call, right?
246 00:24:37.900 ⇒ 00:24:38.450 Samuel Roberts: Right.
247 00:24:38.630 ⇒ 00:24:40.620 Bobby Palmieri: There’s no data warehouse.
248 00:24:40.620 ⇒ 00:24:41.680 Samuel Roberts: Sir.
249 00:24:42.350 ⇒ 00:24:50.089 Bobby Palmieri: when are we… when will we be using the data warehouse versus when do we… like, I want to recreate this in Stitch.
250 00:24:50.090 ⇒ 00:24:51.420 Samuel Roberts: Hmm.
251 00:24:51.420 ⇒ 00:24:55.799 Bobby Palmieri: huge priority, but eventually I want to recreate, you know, something like this in Stitch.
252 00:24:56.220 ⇒ 00:25:01.280 Bobby Palmieri: When do we use the data warehouse versus when do we use the API, and what are kind of the trade-offs there?
253 00:25:04.000 ⇒ 00:25:04.840 Uttam Kumaran: Yeah, same.
254 00:25:04.840 ⇒ 00:25:11.360 Samuel Roberts: I can… we had this conversation internally, because we’re doing something like this for our internal app.
255 00:25:11.390 ⇒ 00:25:21.949 Uttam Kumaran: The MCP is gonna be good for things where you’re pulling, like, individual records, or, like, one number or two numbers. Anything where you have to
256 00:25:22.030 ⇒ 00:25:36.019 Uttam Kumaran: run an aggregation over a lot of information, like, most likely this, they’re not using MCPs for. They’re, they’re, like, they’re, like, pre-aggregating this somewhere, and then displaying the results. But for example, like, pulling, like.
257 00:25:36.020 ⇒ 00:25:43.269 Uttam Kumaran: what is the latest ad ID? That’s, like, a simple API call that we won’t go to the data warehouse for.
258 00:25:43.270 ⇒ 00:25:45.500 Uttam Kumaran: Cause, like, we won’t… yeah.
259 00:25:45.690 ⇒ 00:25:53.950 Bobby Palmieri: Do you think they’re using a data warehouse? Because, like, each time I log in, I have to pull analytics for this specific account. My thought was it’s probably just an API call.
260 00:25:53.950 ⇒ 00:26:02.919 Samuel Roberts: They probably are, yeah. So the difference here is that there’s kind of three things. There’s the MCP, which is how the LLM knows, basically, how to make the API calls.
261 00:26:02.920 ⇒ 00:26:03.630 Bobby Palmieri: Yup.
262 00:26:03.630 ⇒ 00:26:18.479 Samuel Roberts: So it’s doing basically the same thing this is doing, just through this tool of the model context protocol. So it gives these tools, it knows what those tools do, and those tools are basically just making these API calls. So.
263 00:26:18.750 ⇒ 00:26:24.830 Samuel Roberts: We, you know, theoretically, like, the… as long as the access is there, like, this could be pulled, perhaps you could build these.
264 00:26:24.830 ⇒ 00:26:29.540 Uttam Kumaran: It just depends on the sophistication of the amount of data.
265 00:26:29.540 ⇒ 00:26:30.080 Samuel Roberts: I was in fact.
266 00:26:30.080 ⇒ 00:26:49.589 Uttam Kumaran: like, Facebook is not going to give you a cohort model, like, out of the box for the API. So, if it requires post-processing, I… we will do our best to pull as much from the API or MCP as possible, because it’s… it’s quicker, but there will be things that we will have to go to data warehouse to pull if it’s, like.
267 00:26:50.120 ⇒ 00:26:52.000 Uttam Kumaran: For example, modeling across
268 00:26:52.130 ⇒ 00:27:02.999 Uttam Kumaran: like, across brands, things like that. You don’t want… and the alternative… the other fourth option is, like, people will do that, like, in your browser on the fly. That’s, like, again, complicated, too.
269 00:27:03.000 ⇒ 00:27:05.480 Samuel Roberts: That might even be what’s happening here, like, I’m not sure.
270 00:27:05.820 ⇒ 00:27:20.969 Samuel Roberts: If you’re refreshing every time and it took a little while to load there, it could be just doing it, yeah, on the fly, which is doable, depending on how far back you want to go, and how much, you know, but, like, you have to click it every time, and then it’s a little wasteful, so, you know.
271 00:27:20.970 ⇒ 00:27:21.800 Bobby Palmieri: Makes sense.
272 00:27:21.990 ⇒ 00:27:32.400 Samuel Roberts: Yeah. Yeah, there’s options there, and there’s trade-offs to all of them, obviously, so, you know. I think we’re on the right path with, like, the MCPs for the LLM. Having that access is…
273 00:27:32.590 ⇒ 00:27:38.860 Samuel Roberts: Fine, because then if we need to do something more specific with the API, we still can, because we have the right tokens and everything.
274 00:27:39.010 ⇒ 00:27:52.540 Samuel Roberts: But then storing it, using the data warehouse for all the deeper analysis stuff, the cohort stuff, is, I think, what makes sense. Like, if you want a graph like that, like, we could do that, but it’s going to be the same kind of, like, wait for it to load, do all that stuff.
275 00:27:52.540 ⇒ 00:27:53.100 Bobby Palmieri: Cool.
276 00:27:56.630 ⇒ 00:28:01.539 Uttam Kumaran: Okay, great. So tomorrow we have our demo. I think, Sam, you sent some stuff already today.
277 00:28:01.860 ⇒ 00:28:11.750 Samuel Roberts: Yeah, that was just… yeah, so that’s, I didn’t show actually doing all that off, but it’s the same kind of thing I got working with the Facebook. You click through. It’s only my personal account right now, because that’s all.
278 00:28:11.750 ⇒ 00:28:16.850 Bobby Palmieri: What… what do you need to get… our meta account hooked up.
279 00:28:17.050 ⇒ 00:28:19.240 Samuel Roberts: once I get this up into…
280 00:28:19.460 ⇒ 00:28:24.249 Samuel Roberts: staging environment. Yeah. You’ll be able to log in, and so it has…
281 00:28:24.440 ⇒ 00:28:31.760 Samuel Roberts: Basically, the way it’s set up right now is that, one meta account covers everything, And so…
282 00:28:31.940 ⇒ 00:28:46.329 Samuel Roberts: whoever’s logged in will have access to whatever ad accounts that meta account had. So right now, I’m the only one… I’m the only thing I had for Facebook, and so, same with, the Google Ads for the,
283 00:28:46.910 ⇒ 00:28:48.610 Samuel Roberts: Stitch at Lilo Social.
284 00:28:48.610 ⇒ 00:28:49.010 Bobby Palmieri: Yep.
285 00:28:49.010 ⇒ 00:28:55.890 Samuel Roberts: it had that one, so I was able to at least confirm that that’s doing something, but it’s hard for me to go further than that, because I have no data in my…
286 00:28:55.940 ⇒ 00:29:09.870 Samuel Roberts: Facebook meta thing, so. Once we get this up, and I get all the plumbing worked out deployment-wise, we can just clear that, whatever’s in there. It might not even have anything if I haven’t tested it that way, but one of you can test it and
287 00:29:10.040 ⇒ 00:29:18.340 Samuel Roberts: You should be able to see all the accounts, and if there’s not, then that’s a new thing for me to figure out, but from my experience testing it locally so far, it should be alright.
288 00:29:18.790 ⇒ 00:29:19.480 Bobby Palmieri: Awesome.
289 00:29:19.480 ⇒ 00:29:22.849 Samuel Roberts: all the permissions. There might be a couple weird things with the,
290 00:29:23.300 ⇒ 00:29:29.949 Samuel Roberts: like we saw with the Google, you have to get it approved to, like, the next level past testing.
291 00:29:30.230 ⇒ 00:29:34.259 Samuel Roberts: So there might be something like that with Meta as well, but so far, I haven’t encountered that, so…
292 00:29:34.260 ⇒ 00:29:35.350 Bobby Palmieri: Okay, cool.
293 00:29:36.020 ⇒ 00:29:43.399 Uttam Kumaran: And then, Sam, I think maybe it’s… can we talk about, next week, maybe, how… what it’s gonna take for onboarding more people from Lilo?
294 00:29:43.600 ⇒ 00:29:48.910 Uttam Kumaran: like, what is the… what do we need from them to do onboarding, or everybody’s able just to go through Google.
295 00:29:49.110 ⇒ 00:29:58.709 Samuel Roberts: Yeah, ideally, I mean, everyone has a, like, a Google or an email that’s, like, tied to the Google, I’m assuming, or is it two different emails? That’s a good question. Yeah.
296 00:29:58.710 ⇒ 00:29:59.490 Uttam Kumaran: author, I’ll be.
297 00:29:59.490 ⇒ 00:30:14.959 Samuel Roberts: If everyone can just click Google, I’ll probably even get rid of that login with the email and password, if that’s just how you’re gonna do it. Yeah. There might be a few things we need to do to make sure that it’s, like, locked down, so no one’s just creating random accounts, and it’s only your domain and our domain, I guess, for testing.
298 00:30:15.490 ⇒ 00:30:33.760 Samuel Roberts: But it should be really easy to get in there. We don’t have a ton of stuff yet for, like, managing those users, if you want to do more, like, lock people down to only access certain brands or something. That’s a… like, that is very low on… That’s what I figured, I figured. As long as there’s something that we can at least, like.
299 00:30:34.070 ⇒ 00:30:40.310 Bobby Palmieri: My thought is, if we are gonna push that down, we probably just need to be able to search for brands, so you don’t have to…
300 00:30:40.920 ⇒ 00:30:43.769 Samuel Roberts: That’s… yeah, exactly, yeah. And we can… that’s… that’s…
301 00:30:43.990 ⇒ 00:30:48.339 Samuel Roberts: I don’t want to say trivial, because nothing’s completely trivial, but it’s… that’s the closest thing to trivial it could be.
302 00:30:48.340 ⇒ 00:30:57.219 Uttam Kumaran: And then in two weeks, Bobby, I’d like to, once we port over Nano, I’d like for you to just turn off the other version, basically, or, like, restrict access to the other version.
303 00:30:57.220 ⇒ 00:30:59.880 Bobby Palmieri: We’re gonna, yeah, we’re… we’ll… we’ll shut that down.
304 00:30:59.880 ⇒ 00:31:14.179 Uttam Kumaran: So I want to get… make sure everybody, Sam, can log in, and I don’t know, Bob, if you want us to do on… or, like, we can send a loom to everybody, but I just want to verify everybody’s in. We’ll port the new version, and then be like, this is where you go to get it, you know?
305 00:31:14.350 ⇒ 00:31:16.940 Uttam Kumaran: Awesome. So at least now there’s users in there, so…
306 00:31:17.120 ⇒ 00:31:17.910 Bobby Palmieri: Yeah.
307 00:31:18.090 ⇒ 00:31:18.720 Uttam Kumaran: Okay.
308 00:31:18.720 ⇒ 00:31:19.320 Samuel Roberts: I’m good.
309 00:31:20.230 ⇒ 00:31:20.960 Uttam Kumaran: Okay, perfect.
310 00:31:20.960 ⇒ 00:31:25.890 Bobby Palmieri: Yeah, we’re doing a… we’re doing a town hall on Thursday, so a week from today.
311 00:31:26.330 ⇒ 00:31:26.710 Uttam Kumaran: Okay.
312 00:31:27.620 ⇒ 00:31:37.079 Bobby Palmieri: basically walk them through, like, the prototype that Zach and I have, of, like, here’s where we’re going, and then, depending on where we net out,
313 00:31:37.080 ⇒ 00:31:38.909 Uttam Kumaran: Is that in the morning, on Thursday?
314 00:31:39.250 ⇒ 00:31:41.550 Bobby Palmieri: It’s like mid-afternoon.
315 00:31:42.060 ⇒ 00:31:46.609 Uttam Kumaran: Okay, then… then, Sam, let’s see, we’ll see what we can get over the hump by, like…
316 00:31:47.280 ⇒ 00:31:48.560 Uttam Kumaran: Maybe by then.
317 00:31:49.440 ⇒ 00:31:55.529 Samuel Roberts: Yeah, I’m… I’m pretty confident that, like, all the stuff you saw on the loom today should be good to go.
318 00:31:55.930 ⇒ 00:32:05.400 Samuel Roberts: there’s some things I had to merge with, stuff Pranav had done, but that shouldn’t be too crazy. The biggest hurdle right now is just deploying it, and, like, running it locally on my machine, it’s easy to…
319 00:32:05.600 ⇒ 00:32:08.729 Samuel Roberts: Pointed in the right places, but as we get it up into the server.
320 00:32:08.730 ⇒ 00:32:13.890 Uttam Kumaran: I’m gonna go… Sam, I’m gonna go see Pranav later for… for dinner, so I’ll… Oh, nice! …to pep talk.
321 00:32:13.890 ⇒ 00:32:14.969 Samuel Roberts: Alright, sounds good.
322 00:32:15.090 ⇒ 00:32:22.540 Samuel Roberts: But yeah, I’m pretty confident that by Thursday, you know, barring any major issues with, like, Google or Meta access, I think we’re good.
323 00:32:22.960 ⇒ 00:32:26.060 Samuel Roberts: And I think we’re good on that, so… So far.
324 00:32:26.060 ⇒ 00:32:32.140 Uttam Kumaran: So that’s what hopefully we can get… if we can get into a good place where you guys can pull up the actual app in production, and, like, people are good to, like.
325 00:32:32.400 ⇒ 00:32:34.439 Uttam Kumaran: Maybe start logging into stuff.
326 00:32:34.440 ⇒ 00:32:35.070 Samuel Roberts: Yeah.
327 00:32:35.150 ⇒ 00:32:36.410 Uttam Kumaran: That could be solid.
328 00:32:36.700 ⇒ 00:32:37.760 Samuel Roberts: Yeah. Awesome.
329 00:32:39.610 ⇒ 00:32:40.370 Samuel Roberts: Cool.
330 00:32:40.760 ⇒ 00:32:53.699 Uttam Kumaran: Okay, cool. So I’m gonna make some adjustments to, again, kind of spare some stuff down. Bobby and Zach, I’ll DM you, just with, like, budget change. I’m like, and then we can go from there, but I’m gonna… I’m just gonna go ahead, and Sam, we can talk to,
331 00:32:53.940 ⇒ 00:32:57.200 Uttam Kumaran: I would talk to Casey tomorrow, and start to loop him in.
332 00:32:57.700 ⇒ 00:32:58.210 Samuel Roberts: Good.
333 00:32:58.740 ⇒ 00:32:59.480 Uttam Kumaran: Okay.
334 00:33:00.680 ⇒ 00:33:01.380 Samuel Roberts: Perfect.
335 00:33:01.380 ⇒ 00:33:02.280 Uttam Kumaran: Thank you, guys.
336 00:33:02.580 ⇒ 00:33:03.909 Bobby Palmieri: Appreciate the time, guys.
337 00:33:04.060 ⇒ 00:33:05.580 zacfromson: Alright, talk to you later.