Meeting Title: AI Team Retro Date: 2025-09-12 Meeting participants: Samuel Roberts, Casie Aviles, Mustafa Raja
WEBVTT
1 00:00:19.040 ⇒ 00:00:20.150 Samuel Roberts: Blue.
2 00:00:23.620 ⇒ 00:00:24.620 Casie Aviles: Hey, guys.
3 00:00:26.940 ⇒ 00:00:28.080 Samuel Roberts: How we doing today?
4 00:00:28.730 ⇒ 00:00:30.100 Mustafa Raja: Doing good, doing good.
5 00:00:31.460 ⇒ 00:00:31.900 Samuel Roberts: in town.
6 00:00:31.900 ⇒ 00:00:32.800 Mustafa Raja: How are you all?
7 00:00:33.930 ⇒ 00:00:37.149 Samuel Roberts: I’m doing alright. I had some friends in town yesterday.
8 00:00:37.540 ⇒ 00:00:39.979 Samuel Roberts: Which was nice to hang out.
9 00:00:40.230 ⇒ 00:00:44.030 Samuel Roberts: You know, someone who used to live here and moved away and came back to visit, so…
10 00:00:44.720 ⇒ 00:00:48.630 Samuel Roberts: That’s kind of… Just some old places we used to go to, so that was neat.
11 00:00:49.690 ⇒ 00:00:56.020 Casie Aviles: I wish I could go and meet new friends more often, but yeah, we live in separate areas now.
12 00:00:56.420 ⇒ 00:00:58.209 Casie Aviles: It’s kind of hard to meet up.
13 00:00:58.210 ⇒ 00:01:01.639 Samuel Roberts: 100%, that’s… he moved to California from Ohio here.
14 00:01:02.700 ⇒ 00:01:03.410 Casie Aviles: Yeah.
15 00:01:03.970 ⇒ 00:01:07.779 Samuel Roberts: But he came back, partially to, like, meet my new kid and stuff.
16 00:01:10.490 ⇒ 00:01:16.659 Samuel Roberts: Great. Nice. Yeah, cool. How was yesterday after I was out?
17 00:01:18.960 ⇒ 00:01:22.379 Mustafa Raja: Yeah, it was, it was good, it was okay.
18 00:01:24.170 ⇒ 00:01:26.819 Casie Aviles: Yeah, I think for ABC, they…
19 00:01:26.950 ⇒ 00:01:30.979 Casie Aviles: It was good, like, they did, like, the transcripting.
20 00:01:31.450 ⇒ 00:01:32.110 Samuel Roberts: Okay.
21 00:01:32.650 ⇒ 00:01:34.470 Casie Aviles: So yeah, that’s nice.
22 00:01:35.870 ⇒ 00:01:38.369 Casie Aviles: I think, yeah, we’re at a good place then.
23 00:01:38.690 ⇒ 00:01:44.059 Samuel Roberts: Okay, good. Yeah, once I… I was really annoyed that I couldn’t get it quite matching,
24 00:01:45.190 ⇒ 00:01:55.299 Samuel Roberts: But I think… I think I’m just… I’m pretty close to getting it to actually finding the Andy. You know, it’s a… I just gotta match them up somehow. And parsing through those timestamps is a little annoying, so I’m wondering if there’s a way…
25 00:01:56.370 ⇒ 00:02:01.429 Samuel Roberts: I’m sure there’s a way. I’m wondering if it’s worth running a script on that to parse the timestamps for all of them?
26 00:02:02.370 ⇒ 00:02:03.850 Casie Aviles: Yeah, yeah.
27 00:02:04.270 ⇒ 00:02:05.140 Casie Aviles: should do that.
28 00:02:05.530 ⇒ 00:02:10.500 Samuel Roberts: And then update, maybe, the saving to Snowflake to, like, also do a parsing step?
29 00:02:12.450 ⇒ 00:02:12.839 Casie Aviles: I’m going.
30 00:02:12.840 ⇒ 00:02:16.319 Samuel Roberts: forward will have the timestamps, because that’ll make querying it a lot easier.
31 00:02:17.690 ⇒ 00:02:20.590 Casie Aviles: Okay, yeah, yeah, definitely, we should do that.
32 00:02:20.590 ⇒ 00:02:28.219 Samuel Roberts: So, yeah, I just, obviously, I didn’t really have time to get into that yesterday, so I was just like, let me just figure out if I can do this the quick and dirty way, and I got close, but not quite there.
33 00:02:28.570 ⇒ 00:02:35.629 Samuel Roberts: But… Besides that, yeah, I was excited I got, at least as far as I got.
34 00:02:36.020 ⇒ 00:02:38.749 Samuel Roberts: And it’s so amazing what Cursor can do, because, like.
35 00:02:39.190 ⇒ 00:02:43.659 Samuel Roberts: I was just, like, testing the API, and I gave it a bunch of routes from the docs, and…
36 00:02:43.660 ⇒ 00:02:49.169 Casie Aviles: They explored the 8x8 API with me, and that’s kind of how that whole UI came to be, was just me exploring the API.
37 00:02:49.410 ⇒ 00:02:55.449 Samuel Roberts: So basically, like, Cursed let me build, like, a custom tool for testing specifically 8x8, which is neat.
38 00:02:56.360 ⇒ 00:02:56.920 Samuel Roberts: What?
39 00:02:56.920 ⇒ 00:02:57.640 Casie Aviles: Nice.
40 00:02:57.640 ⇒ 00:03:01.529 Samuel Roberts: That would have taken me so long without AI, you know, like, 5 years ago.
41 00:03:03.010 ⇒ 00:03:07.209 Casie Aviles: Yeah, like, you’d have to go through all the docs, and you’d have to test, like.
42 00:03:07.210 ⇒ 00:03:15.139 Samuel Roberts: Yeah, and even with the docs, I was feeding the docs into Cursor, and it was helping me, like, because the API key only has access to certain things, and I didn’t know what was turned on and stuff.
43 00:03:15.360 ⇒ 00:03:16.120 Casie Aviles: Hmm.
44 00:03:16.280 ⇒ 00:03:18.469 Samuel Roberts: So I was just, like, testing everything I could test.
45 00:03:20.670 ⇒ 00:03:24.579 Samuel Roberts: It was pretty good, pretty good. But, yeah, anyway,
46 00:03:26.040 ⇒ 00:03:30.090 Samuel Roberts: I guess, let’s… yeah, maybe, like, the little… a quick retro,
47 00:03:30.260 ⇒ 00:03:34.010 Samuel Roberts: What else was good, what else was bad today, like, er, this week?
48 00:03:35.690 ⇒ 00:03:39.599 Samuel Roberts: Other client work, what do you guys, you know, how are you guys feeling about this week?
49 00:03:43.310 ⇒ 00:03:45.129 Casie Aviles: Once I got, I guess I’ll…
50 00:03:45.560 ⇒ 00:03:47.449 Samuel Roberts: Yeah, yeah, go ahead, Mustafa.
51 00:03:47.470 ⇒ 00:03:48.330 Casie Aviles: If you want them.
52 00:03:48.330 ⇒ 00:03:58.820 Mustafa Raja: Yeah, so for me, I guess it was, it was, Jill, I guess. Yeah.
53 00:03:59.440 ⇒ 00:04:09.440 Mustafa Raja: We started in Tedude this week, and we have some progress on the milestones we had set for this week.
54 00:04:09.480 ⇒ 00:04:21.049 Mustafa Raja: Apart from the evals, because it tries to, access Git, in, Dagster, and Dagster environment doesn’t have Git installed.
55 00:04:21.870 ⇒ 00:04:22.570 Samuel Roberts: Oh, interesting.
56 00:04:22.570 ⇒ 00:04:39.269 Mustafa Raja: I need to figure that out. Apart from that, everything looks good on that side. On default, I believe I was able to, conclude the tiers, work so far that we have, planned.
57 00:04:40.430 ⇒ 00:04:46.319 Mustafa Raja: And so maybe today I can move on to add more signals,
58 00:04:46.370 ⇒ 00:05:04.080 Mustafa Raja: to Salesforce. They did move, all of the… all of their accounts, the 40,000 accounts that they have in their Salesforce, to the Brainforge list, so I won’t be able to test that, so I’ll have to either create a new list, meaning a new property.
59 00:05:04.390 ⇒ 00:05:10.670 Mustafa Raja: Yeah. Oh, let’s see how… how that looks like, I’m not sure for now.
60 00:05:10.670 ⇒ 00:05:12.130 Samuel Roberts: Definitely. Okay.
61 00:05:14.280 ⇒ 00:05:15.300 Samuel Roberts: Interesting.
62 00:05:16.470 ⇒ 00:05:20.240 Samuel Roberts: Yeah, so you may need to keep track some other way, I see what you’re saying. Okay.
63 00:05:21.650 ⇒ 00:05:29.869 Samuel Roberts: Alright, good. And the… the Salesforce stuff with them, they want it on hold, right?
64 00:05:30.060 ⇒ 00:05:30.920 Samuel Roberts: For the triggering?
65 00:05:30.920 ⇒ 00:05:38.260 Mustafa Raja: Yeah, yeah, yeah, they said the… So Ryan said that he wants to dig a little deeper in that.
66 00:05:38.390 ⇒ 00:05:45.270 Mustafa Raja: And later he said that let’s put some hold onto it, maybe he wants to talk about it first.
67 00:05:45.310 ⇒ 00:05:46.659 Samuel Roberts: Okay, yeah, I figured that makes sense.
68 00:05:46.660 ⇒ 00:05:47.290 Mustafa Raja: forward.
69 00:05:47.560 ⇒ 00:05:48.190 Samuel Roberts: Yeah.
70 00:05:49.500 ⇒ 00:05:50.300 Samuel Roberts: Okay.
71 00:05:51.750 ⇒ 00:05:52.700 Samuel Roberts: Casey?
72 00:05:54.200 ⇒ 00:05:58.290 Casie Aviles: Okay, I think for this week, it was…
73 00:05:59.490 ⇒ 00:06:04.149 Casie Aviles: I’m just at a pretty steady pace of… Getting things out.
74 00:06:05.410 ⇒ 00:06:07.190 Casie Aviles: Nothing too crazy, I think.
75 00:06:07.380 ⇒ 00:06:16.319 Casie Aviles: This week was more… Yeah, it’s a lot less hectic than some past weeks before.
76 00:06:16.640 ⇒ 00:06:23.910 Casie Aviles: Yeah, I think I was able to make some steady progress on both Antonio and ABC.
77 00:06:25.120 ⇒ 00:06:28.239 Casie Aviles: Hmm, what else? I think… yeah, I guess…
78 00:06:29.000 ⇒ 00:06:32.890 Casie Aviles: For just, for both, there are some things that…
79 00:06:33.330 ⇒ 00:06:39.260 Casie Aviles: maybe I… might have overlooked, like, for both the database stuff and also the…
80 00:06:39.670 ⇒ 00:06:45.870 Casie Aviles: what do you call this? The automation, so… maybe, like…
81 00:06:46.840 ⇒ 00:06:54.200 Casie Aviles: Well, it would be great if, you know, I could… I could also, like, get some…
82 00:06:54.640 ⇒ 00:06:56.150 Casie Aviles: Reviews on that.
83 00:06:56.860 ⇒ 00:06:57.500 Samuel Roberts: Okay.
84 00:06:57.800 ⇒ 00:06:59.369 Samuel Roberts: Yeah, like… Yeah.
85 00:06:59.370 ⇒ 00:07:04.170 Casie Aviles: Once I’ve wrapped up all this stuff, but yeah, I think…
86 00:07:04.780 ⇒ 00:07:08.249 Casie Aviles: We’re very close to just, you know, getting all of it.
87 00:07:08.820 ⇒ 00:07:09.760 Casie Aviles: Done.
88 00:07:10.710 ⇒ 00:07:13.499 Casie Aviles: And yeah, I’m excited for ABC, like.
89 00:07:13.780 ⇒ 00:07:20.710 Casie Aviles: I guess we have some next steps, like actually getting it connected to Andy, and then… Also, implementing the…
90 00:07:21.350 ⇒ 00:07:30.480 Casie Aviles: what do you call this? The update flow, which I talked with Oish, so we discussed how we could… how we could implement it.
91 00:07:30.690 ⇒ 00:07:31.300 Samuel Roberts: Okay, cool.
92 00:07:31.300 ⇒ 00:07:32.450 Casie Aviles: I think, yeah.
93 00:07:33.970 ⇒ 00:07:37.250 Samuel Roberts: Great, yeah, that sounds like good progress this week, and
94 00:07:38.450 ⇒ 00:07:40.529 Samuel Roberts: I’m glad we’re moving forward with some of that stuff.
95 00:07:41.050 ⇒ 00:07:43.520 Samuel Roberts: That’s cool.
96 00:07:44.140 ⇒ 00:07:50.400 Samuel Roberts: I guess let’s just jump into Cursor real quick, too, and just see what’s on for today, but…
97 00:07:50.620 ⇒ 00:07:53.679 Samuel Roberts: or not cursor, sorry, layer.
98 00:07:54.230 ⇒ 00:07:56.189 Samuel Roberts: I got cursor on the brain.
99 00:07:58.490 ⇒ 00:08:02.430 Samuel Roberts: So I guess let’s, like, go see what we’re finishing up today, hopefully, and…
100 00:08:04.220 ⇒ 00:08:06.490 Casie Aviles: E, or ABCs.
101 00:08:07.760 ⇒ 00:08:12.079 Casie Aviles: I think there are some triage tickets that I haven’t looked at yet.
102 00:08:12.500 ⇒ 00:08:13.340 Samuel Roberts: Yeah.
103 00:08:15.160 ⇒ 00:08:18.849 Samuel Roberts: Alright, good, and I got this going, I should probably add an update here or two.
104 00:08:20.110 ⇒ 00:08:21.270 Samuel Roberts: Oh, okay.
105 00:08:22.800 ⇒ 00:08:29.689 Samuel Roberts: We’ll jump back to these after. The default, kind of just had that update, but anything specific today?
106 00:08:33.179 ⇒ 00:08:34.089 Mustafa Raja: Yeah.
107 00:08:34.439 ⇒ 00:08:40.249 Mustafa Raja: I have moved a lot of stuff in the… Internal review.
108 00:08:40.529 ⇒ 00:08:44.149 Samuel Roberts: Okay. Yeah, okay.
109 00:08:44.150 ⇒ 00:08:46.580 Mustafa Raja: I guess,
110 00:08:46.710 ⇒ 00:08:54.529 Mustafa Raja: We can improve whatever the implementation I have implemented over here once the block ticket gets unblocked.
111 00:08:55.180 ⇒ 00:08:55.730 Samuel Roberts: Okay.
112 00:08:55.770 ⇒ 00:09:15.359 Mustafa Raja: So, what this is really blocked by is the source of data. We do not have it. We want to do PDL people’s data labs, and the API key… the API key that default people gave us,
113 00:09:16.540 ⇒ 00:09:21.290 Mustafa Raja: it only has a few fields, enabled.
114 00:09:21.740 ⇒ 00:09:22.339 Samuel Roberts: Okay.
115 00:09:22.340 ⇒ 00:09:30.800 Mustafa Raja: now… That’s all good because it covers most of it. It only excludes one signal.
116 00:09:31.310 ⇒ 00:09:36.060 Mustafa Raja: And the signal is that, if the…
117 00:09:36.060 ⇒ 00:09:52.759 Mustafa Raja: sales people in a company are spread across the globe or not. Meaning if they are working across, more than 3 countries or not. And for that, we’ll have… we would need, the employees’ locations.
118 00:09:52.820 ⇒ 00:09:54.820 Mustafa Raja: To be able to infer that.
119 00:09:55.010 ⇒ 00:10:03.489 Mustafa Raja: Yeah. And this is something that they do not have with, in their deal with People’s Data Lab, so let’s.
120 00:10:03.860 ⇒ 00:10:14.630 Mustafa Raja: I have pinged them and, explained to them what’s happening, and, I was testing, yesterday and, or the day before yesterday.
121 00:10:15.700 ⇒ 00:10:18.829 Mustafa Raja: And they ran out of credits.
122 00:10:18.830 ⇒ 00:10:19.360 Samuel Roberts: Oh, gotcha.
123 00:10:19.360 ⇒ 00:10:25.249 Mustafa Raja: So, for that sort of thing, they only have 100 credits, does not, include in the degree.
124 00:10:25.860 ⇒ 00:10:26.870 Samuel Roberts: Got it, okay.
125 00:10:27.360 ⇒ 00:10:35.510 Mustafa Raja: Yeah, so I haven’t heard back from them, since yesterday, so let’s see. I’m hoping to get a response today.
126 00:10:36.440 ⇒ 00:10:46.650 Mustafa Raja: So once that data will be available, I guess, the overall quality of our, system, sorry, signals, will also improve.
127 00:10:47.340 ⇒ 00:10:48.320 Samuel Roberts: Okay, great.
128 00:10:48.460 ⇒ 00:10:52.470 Mustafa Raja: One, one more thing, happened yesterday,
129 00:10:52.780 ⇒ 00:10:56.459 Mustafa Raja: I can explain it, if I… if I can share my screen.
130 00:10:56.550 ⇒ 00:10:57.490 Samuel Roberts: Sure, yeah.
131 00:11:00.410 ⇒ 00:11:02.710 Mustafa Raja: Sorry, I sent request.
132 00:11:02.830 ⇒ 00:11:03.430 Mustafa Raja: Yeah.
133 00:11:03.430 ⇒ 00:11:04.589 Samuel Roberts: I’m good, okay again.
134 00:11:05.150 ⇒ 00:11:08.540 Mustafa Raja: Okay, so what happened was…
135 00:11:15.310 ⇒ 00:11:30.190 Mustafa Raja: Okay, what happened was, I saw that, I was working with the, I was working in Clay, and I saw that, almost 40,000 accounts are being pushed to Clay, tables.
136 00:11:31.090 ⇒ 00:11:48.429 Mustafa Raja: And that’s a lot. That’s going to cost them all of their credits that they have. So I, blocked the run, and I, informed with them about this. We did, verify with them, and it was them. Now,
137 00:11:48.650 ⇒ 00:11:57.470 Mustafa Raja: I had to work on Clay, so I didn’t immediately, restore the connections with, what’s it called?
138 00:11:58.150 ⇒ 00:12:07.349 Mustafa Raja: Salesforce. But now that I… now that it’s done, and I want to restore the connection,
139 00:12:07.840 ⇒ 00:12:09.960 Mustafa Raja: Sorry,
140 00:12:14.750 ⇒ 00:12:20.339 Mustafa Raja: So this is, this is only for one table. This is the cost to run only one table,
141 00:12:20.340 ⇒ 00:12:21.970 Samuel Roberts: No, this is about…
142 00:12:21.970 ⇒ 00:12:32.749 Mustafa Raja: And they only have 10% of these credits, meaning they have 36K, and this is going to take 300… 363K credits,
143 00:12:33.220 ⇒ 00:12:38.789 Mustafa Raja: And, so we can see that they have only 38,000 left.
144 00:12:38.790 ⇒ 00:12:39.840 Samuel Roberts: Yeah, okay.
145 00:12:39.840 ⇒ 00:12:43.800 Mustafa Raja: So… Yeah, what should I do? I mean.
146 00:12:43.800 ⇒ 00:12:44.399 Samuel Roberts: Yeah, awesome.
147 00:12:44.400 ⇒ 00:12:46.960 Mustafa Raja: We stored the connection, right now, but.
148 00:12:46.960 ⇒ 00:12:47.839 Samuel Roberts: You see that?
149 00:12:47.840 ⇒ 00:12:51.369 Mustafa Raja: 48… 40,000… Rows are pulled.
150 00:12:51.550 ⇒ 00:12:53.120 Samuel Roberts: Yeah, okay.
151 00:12:54.160 ⇒ 00:12:56.960 Samuel Roberts: That is a really good question. So…
152 00:12:58.000 ⇒ 00:13:06.719 Mustafa Raja: Yeah, I just need to… I just need to connect the website, column with this, and it’s… this table is going to run, but this table needs 300,000 credits.
153 00:13:06.720 ⇒ 00:13:10.900 Samuel Roberts: Okay, yeah. I guess I’m… I’m wondering…
154 00:13:11.670 ⇒ 00:13:15.220 Samuel Roberts: Yeah, they… so they need to get those credits if they want to run it, basically, so they need.
155 00:13:15.220 ⇒ 00:13:25.270 Mustafa Raja: Yeah, yeah. I mean, I can establish the connection right now, and it’s going to consume the credits, but still, there won’t be any results though, right?
156 00:13:25.750 ⇒ 00:13:31.660 Samuel Roberts: Right, I would say don’t do that, and we want to confirm, because if they can get… if they can set the credits up.
157 00:13:31.760 ⇒ 00:13:33.700 Samuel Roberts: Then it’s worth doing, you know?
158 00:13:34.140 ⇒ 00:13:42.660 Mustafa Raja: Yeah, yeah, yeah. Because then, we’ll lose this. We… we did lose, I believe we did lose,
159 00:13:43.030 ⇒ 00:13:46.030 Mustafa Raja: 20,000… no.
160 00:13:46.030 ⇒ 00:13:46.460 Samuel Roberts: Okay.
161 00:13:46.460 ⇒ 00:13:47.500 Mustafa Raja: 30,000.
162 00:13:48.100 ⇒ 00:13:50.880 Mustafa Raja: Yeah, around 30,000 credits.
163 00:13:51.220 ⇒ 00:13:54.179 Mustafa Raja: Because one of these… one of these tables did run.
164 00:13:54.480 ⇒ 00:13:56.299 Samuel Roberts: With all the… with all 40, or 30.
165 00:13:56.300 ⇒ 00:14:09.800 Mustafa Raja: Yeah, yeah, yeah. Yeah, but the cost for that table was only 1 credit per row, so that was pretty affordable. This table is 9 credits per row.
166 00:14:10.930 ⇒ 00:14:11.520 Samuel Roberts: Alright.
167 00:14:11.750 ⇒ 00:14:14.799 Samuel Roberts: I have no idea how much these credits cost, but I…
168 00:14:14.800 ⇒ 00:14:18.410 Mustafa Raja: Yeah, me too. I don’t… I don’t have any idea, too.
169 00:14:18.410 ⇒ 00:14:25.119 Samuel Roberts: Yeah, let me see real quick, make sure, because I want to have a sense of, like, how much money we’re talking here for them, because I feel like they should still approve it, because they need to…
170 00:14:25.410 ⇒ 00:14:31.099 Mustafa Raja: Yeah, I mean, it was them who, added those accounts to…
171 00:14:31.460 ⇒ 00:14:32.730 Samuel Roberts: Yeah, exactly.
172 00:14:32.730 ⇒ 00:14:39.300 Mustafa Raja: Right? So they know about this. I don’t think they have idea about the credits it’s going to consume.
173 00:14:39.960 ⇒ 00:14:42.530 Samuel Roberts: Yeah, no, I definitely need to let them know.
174 00:14:42.820 ⇒ 00:14:45.100 Samuel Roberts: Cause I’m looking at it here.
175 00:14:46.110 ⇒ 00:14:49.350 Samuel Roberts: It is… a lot of money, it looks like.
176 00:14:49.800 ⇒ 00:14:50.310 Mustafa Raja: Oh.
177 00:14:50.310 ⇒ 00:14:55.890 Samuel Roberts: $2,000 a month is… Yeah,
178 00:14:56.980 ⇒ 00:15:01.419 Samuel Roberts: Like, the only way to get those credits, unless they have a deal or something that we don’t know about.
179 00:15:02.110 ⇒ 00:15:04.820 Mustafa Raja: Yeah, I think they do have them, right?
180 00:15:04.820 ⇒ 00:15:11.490 Samuel Roberts: Yeah, they must have something to get, so maybe someone on their team needs to figure that out, and decide if they really want to run all of these, you know?
181 00:15:12.880 ⇒ 00:15:13.680 Mustafa Raja: Yeah, yeah, yeah.
182 00:15:15.490 ⇒ 00:15:24.179 Mustafa Raja: I was able to catch it, in time, so that only one table was able to run, not all.
183 00:15:24.380 ⇒ 00:15:25.070 Samuel Roberts: Yeah.
184 00:15:25.330 ⇒ 00:15:29.859 Mustafa Raja: And I was able to close the rest.
185 00:15:30.140 ⇒ 00:15:33.630 Mustafa Raja: On 520, on 550, approximately.
186 00:15:33.630 ⇒ 00:15:37.419 Samuel Roberts: So that’s… that’s just one table you said, right? Is that how, like, how many tables are we doing now?
187 00:15:37.420 ⇒ 00:15:43.899 Mustafa Raja: So, this one table, this first table that we have, this is a combination of three, three signals.
188 00:15:44.350 ⇒ 00:15:44.800 Samuel Roberts: Got it.
189 00:15:44.800 ⇒ 00:15:47.330 Mustafa Raja: And this table is going to cost a lot.
190 00:15:47.740 ⇒ 00:15:50.679 Samuel Roberts: Yeah, are the other tables all less than that, you think?
191 00:15:51.200 ⇒ 00:16:02.860 Mustafa Raja: Yeah, the other tables are going to be less than 9, but, I get, for now, till now, it’s closer to 19 credits per company.
192 00:16:03.870 ⇒ 00:16:04.850 Samuel Roberts: Got it, okay.
193 00:16:04.990 ⇒ 00:16:07.000 Mustafa Raja: So, we have 40,000 companies…
194 00:16:07.000 ⇒ 00:16:07.340 Samuel Roberts: Yeah.
195 00:16:07.340 ⇒ 00:16:13.179 Mustafa Raja: And each company is going to take 19 credits, so we can multiply that.
196 00:16:13.180 ⇒ 00:16:13.600 Samuel Roberts: Yeah.
197 00:16:13.600 ⇒ 00:16:15.980 Mustafa Raja: These two figures to get the credits that we would need.
198 00:16:16.170 ⇒ 00:16:19.659 Samuel Roberts: Yeah, 760,000 credits, which is…
199 00:16:20.760 ⇒ 00:16:21.810 Mustafa Raja: A lot of credits.
200 00:16:21.810 ⇒ 00:16:27.570 Samuel Roberts: a lot of credits based on what I’m seeing here. I didn’t really know how this worked, and a lot of money. Okay, let’s,
201 00:16:32.330 ⇒ 00:16:34.670 Samuel Roberts: I would say let’s let them know that.
202 00:16:36.030 ⇒ 00:16:37.649 Samuel Roberts: And…
203 00:16:37.650 ⇒ 00:16:44.620 Mustafa Raja: Apart from credits also, browser-based will also fail.
204 00:16:46.100 ⇒ 00:16:46.880 Samuel Roberts: Oh.
205 00:16:47.140 ⇒ 00:16:48.529 Mustafa Raja: Let’s open that up.
206 00:16:48.830 ⇒ 00:16:49.580 Samuel Roberts: Yeah.
207 00:16:54.320 ⇒ 00:16:59.349 Mustafa Raja: Sorry, my internet is a little… On the slower side…
208 00:17:00.220 ⇒ 00:17:01.899 Mustafa Raja: It was because of the rains.
209 00:17:02.450 ⇒ 00:17:06.429 Mustafa Raja: Develop a plan…
210 00:17:14.450 ⇒ 00:17:19.010 Mustafa Raja: Yeah, so we see that we already have 25 hours consumed, right?
211 00:17:19.829 ⇒ 00:17:20.189 Samuel Roberts: Okay.
212 00:17:20.190 ⇒ 00:17:25.490 Mustafa Raja: and I, and this is,
213 00:17:26.010 ⇒ 00:17:30.090 Mustafa Raja: Developer plan, so let’s see what developer plan includes.
214 00:17:33.710 ⇒ 00:17:47.880 Mustafa Raja: Developer plan includes 100 browser hours. So there’s no way that, on 500 accounts, we were… we… we hit 25 hours, and 40,000 accounts, we’ll be able to cover all of this in this.
215 00:17:48.490 ⇒ 00:17:53.249 Samuel Roberts: And then it’s how much… it’s 12 cents per browser hour after that?
216 00:17:54.070 ⇒ 00:17:55.180 Samuel Roberts: Yeah. Okay.
217 00:17:56.190 ⇒ 00:18:00.090 Samuel Roberts: Alright, is this our browser-based account?
218 00:18:00.210 ⇒ 00:18:01.359 Mustafa Raja: Nope, this is good.
219 00:18:01.360 ⇒ 00:18:02.520 Samuel Roberts: Theirs? Okay.
220 00:18:02.520 ⇒ 00:18:02.980 Mustafa Raja: Yeah, yeah, yeah.
221 00:18:02.980 ⇒ 00:18:08.490 Samuel Roberts: So, I’m thinking we need to… Put a little,
222 00:18:08.640 ⇒ 00:18:10.960 Samuel Roberts: Like, spreadsheet together, maybe, for them?
223 00:18:12.370 ⇒ 00:18:12.890 Samuel Roberts: That was.
224 00:18:12.890 ⇒ 00:18:13.280 Mustafa Raja: Hmm.
225 00:18:13.280 ⇒ 00:18:15.390 Samuel Roberts: Joe, like, you know.
226 00:18:15.630 ⇒ 00:18:20.049 Samuel Roberts: This one table’s gonna cost this month, and then these are 19, so we know it’s gonna be this.
227 00:18:20.170 ⇒ 00:18:22.670 Mustafa Raja: Yeah, yeah, after that… Sorry.
228 00:18:22.910 ⇒ 00:18:23.550 Samuel Roberts: Go ahead.
229 00:18:23.890 ⇒ 00:18:26.029 Mustafa Raja: Yeah, after that, there’s 2 more coasts.
230 00:18:26.360 ⇒ 00:18:31.190 Mustafa Raja: Okay. One is for their stack?
231 00:18:32.810 ⇒ 00:18:33.390 Samuel Roberts: Okay.
232 00:18:33.390 ⇒ 00:18:41.639 Mustafa Raja: And this is our API. And this gives us free, sorry, fixed credits, so we…
233 00:18:41.840 ⇒ 00:18:47.080 Mustafa Raja: We would not incur any cost, it will just stop running once we hit our quota.
234 00:18:47.280 ⇒ 00:18:51.580 Mustafa Raja: The other thing is that we, we have
235 00:18:51.990 ⇒ 00:18:58.789 Mustafa Raja: connected their, chat, sorry, OpenAI key.
236 00:18:58.790 ⇒ 00:19:00.970 Samuel Roberts: Oh, right, for browser phase. Yes, you’re right.
237 00:19:00.970 ⇒ 00:19:02.960 Mustafa Raja: So that is going to cost 2.
238 00:19:06.640 ⇒ 00:19:07.800 Samuel Roberts: Yes. So we have, like.
239 00:19:07.800 ⇒ 00:19:10.560 Mustafa Raja: Multiple, multiple, costs.
240 00:19:11.280 ⇒ 00:19:14.579 Samuel Roberts: Do we have a sense of how much per run the…
241 00:19:14.580 ⇒ 00:19:24.489 Mustafa Raja: Yeah, yeah, yeah, I, I did, a few days ago, check, the… what’s it called? The OpenAI cost per…
242 00:19:25.250 ⇒ 00:19:31.419 Mustafa Raja: company, or per browser-based run, and that did come around 3 cents.
243 00:19:31.830 ⇒ 00:19:37.650 Samuel Roberts: 3 cents? Okay. So I think we just need to type all this out, like, on a per row basis, right?
244 00:19:37.910 ⇒ 00:19:38.320 Mustafa Raja: Yeah.
245 00:19:38.320 ⇒ 00:19:44.560 Samuel Roberts: Or per company basis, I guess. So we have 19 credits per company, Right, we…
246 00:19:44.560 ⇒ 00:19:49.919 Mustafa Raja: We can actually double-check that to verify if it’s actually 19 or less.
247 00:19:49.920 ⇒ 00:19:50.989 Samuel Roberts: Okay, yeah.
248 00:19:51.350 ⇒ 00:19:58.490 Mustafa Raja: Let’s do this… So, 0.2 for this one…
249 00:19:58.490 ⇒ 00:19:59.549 Samuel Roberts: Oh, I see, okay.
250 00:20:03.310 ⇒ 00:20:11.650 Mustafa Raja: 3 for this one, 12 for this one… 9 for this one… Yeah, hmm.
251 00:20:12.080 ⇒ 00:20:14.250 Mustafa Raja: We aren’t using this one.
252 00:20:14.650 ⇒ 00:20:20.680 Mustafa Raja: This one isn’t being used. So 9 and 12 is 21, I guess, right?
253 00:20:21.100 ⇒ 00:20:21.690 Samuel Roberts: Yep.
254 00:20:22.440 ⇒ 00:20:28.710 Mustafa Raja: Copy of routing enrichment. I think we are not using that, let me… We’ll verify.
255 00:20:31.540 ⇒ 00:20:32.630 Mustafa Raja: enrichment.
256 00:20:35.360 ⇒ 00:20:41.069 Mustafa Raja: Sunny… Hmm… I think we’re using it.
257 00:20:41.730 ⇒ 00:20:42.200 Samuel Roberts: Okay.
258 00:20:43.660 ⇒ 00:20:44.960 Mustafa Raja: Okay, let’s go back.
259 00:20:47.310 ⇒ 00:20:48.010 Mustafa Raja: Hmm.
260 00:20:51.750 ⇒ 00:21:02.920 Mustafa Raja: From here… so 9… 21, 21, 22… 25N… 25.2.
261 00:21:04.140 ⇒ 00:21:09.120 Samuel Roberts: 25.2 per… okay. Let’s, I’m gonna open up a…
262 00:21:09.350 ⇒ 00:21:13.700 Samuel Roberts: I think we need to let them know, just so they know, before we run it all, right?
263 00:21:13.700 ⇒ 00:21:15.350 Mustafa Raja: I think it’s more than 25.
264 00:21:15.590 ⇒ 00:21:23.129 Samuel Roberts: Okay. I’ll have to thoroughly go through this again. Okay, yeah, I would say let’s make a, like, a quick spreadsheet.
265 00:21:23.410 ⇒ 00:21:24.840 Samuel Roberts: Just in Google Sheets.
266 00:21:25.620 ⇒ 00:21:30.950 Samuel Roberts: And have, like, here’s the clay costs, and then you can just sum them up, right?
267 00:21:31.630 ⇒ 00:21:33.529 Mustafa Raja: And then we know that’s per…
268 00:21:33.530 ⇒ 00:21:37.839 Samuel Roberts: per company. Then, let’s pull in the browser-based info.
269 00:21:38.230 ⇒ 00:21:41.380 Samuel Roberts: the… ChatGPT estimate.
270 00:21:42.060 ⇒ 00:21:46.350 Samuel Roberts: And then I guess we’re gonna have to do the… what was it, the there stack?
271 00:21:47.760 ⇒ 00:21:51.929 Mustafa Raja: Their stack, we are using, our API key for that, so…
272 00:21:51.930 ⇒ 00:21:54.370 Samuel Roberts: Right, but we’re gonna have to up that, aren’t we? So…
273 00:21:54.580 ⇒ 00:21:56.719 Mustafa Raja: Yeah, yeah, yeah, to pull all of this.
274 00:21:57.340 ⇒ 00:21:57.920 Mustafa Raja: we’ll have.
275 00:21:57.920 ⇒ 00:21:58.840 Samuel Roberts: Yeah.
276 00:21:58.840 ⇒ 00:21:59.310 Mustafa Raja: Up, down.
277 00:21:59.310 ⇒ 00:22:11.060 Samuel Roberts: So, let’s just… can you just get that, like, maybe even on a spreadsheet, or just, like, in a Notion doc or something, just get all these numbers and, like, links to pricing together, so we can maybe do some quick estimates?
278 00:22:12.410 ⇒ 00:22:13.379 Mustafa Raja: Yeah, yeah, yeah.
279 00:22:13.590 ⇒ 00:22:22.560 Samuel Roberts: And then we can give them a, like… we can run all 40,000 of those, but it’s gonna have to… we’re gonna have to upgrade these 3 accounts and pay more, you know, here.
280 00:22:22.840 ⇒ 00:22:28.199 Samuel Roberts: Just to make sure they know, because I would feel bad running this, and then them being like, oh, we didn’t realize it was going to be so much.
281 00:22:28.630 ⇒ 00:22:31.769 Samuel Roberts: Or they’re fine with it. I don’t know what their funding is like, you know?
282 00:22:32.360 ⇒ 00:22:36.710 Mustafa Raja: Yeah, same. That is why, I stopped it.
283 00:22:36.990 ⇒ 00:22:46.829 Mustafa Raja: Yeah, that’s perfect. I definitely think, yeah. One, I stopped it because it’s not going to run at all, to be honest, because 300,000 is going to be consumed by this one only.
284 00:22:46.830 ⇒ 00:22:50.280 Samuel Roberts: Yeah, exactly, exactly. Okay. So yeah, let’s,
285 00:22:50.400 ⇒ 00:23:08.750 Samuel Roberts: Yeah, I would say put together, like, a Notion doc, or a Google spreadsheet, whatever you feel more comfortable in, and just, like, get all of these things that we just talked about down on paper, for lack of a better word, and then we can try to get, like, a total price on, like, a… when you put a company in, it’s gonna cost…
286 00:23:08.850 ⇒ 00:23:12.180 Samuel Roberts: Estimate of how much, and then we can do that out to $40,000.
287 00:23:13.970 ⇒ 00:23:22.799 Mustafa Raja: So, so, one thing, I, I would want to, clarify, or would want your advice on, for the,
288 00:23:22.900 ⇒ 00:23:37.060 Mustafa Raja: API key that we have for OpenAI, what I did was, I, calculated an average, for each, each run. So, so is this something acceptable?
289 00:23:38.000 ⇒ 00:23:45.350 Samuel Roberts: I think for now, until we have a better insight, so, on browser base, how is that being called?
290 00:23:45.910 ⇒ 00:23:46.560 Samuel Roberts: Nothing.
291 00:23:46.560 ⇒ 00:23:48.000 Mustafa Raja: Oh, I have a script.
292 00:23:48.220 ⇒ 00:23:52.400 Samuel Roberts: Yeah, we have it on our, Heroku.
293 00:23:53.100 ⇒ 00:23:56.179 Samuel Roberts: Yes, okay, so I think what we could probably do…
294 00:23:56.340 ⇒ 00:24:02.210 Samuel Roberts: is I’ve been exploring some instrumentation stuff, so if we want a more accurate reading.
295 00:24:02.660 ⇒ 00:24:10.020 Samuel Roberts: We would have to wrap that with some instrumentation around the OpenAI call. And the question is, does browser-based more easily allow us to do that or not?
296 00:24:11.080 ⇒ 00:24:20.340 Mustafa Raja: Yeah, I guess we, we are… the… the… what’s it called? The key is already wrapped in a, stage hand…
297 00:24:20.500 ⇒ 00:24:22.060 Mustafa Raja: Rapper.
298 00:24:22.250 ⇒ 00:24:29.759 Mustafa Raja: Yeah, so that’s a good question. We’ll be able to do that, with that. Let me pull up the code, actually.
299 00:24:30.930 ⇒ 00:24:32.310 Samuel Roberts: Yeah, that’s a good question.
300 00:24:35.860 ⇒ 00:24:36.940 Mustafa Raja: Stay’s hand.
301 00:24:40.450 ⇒ 00:24:41.530 Mustafa Raja: Oblanky.
302 00:24:42.120 ⇒ 00:24:48.509 Mustafa Raja: Yeah, see, so this is the stage hand configuration that we are using here in this…
303 00:24:48.510 ⇒ 00:24:49.020 Samuel Roberts: Okay.
304 00:24:49.020 ⇒ 00:24:58.559 Mustafa Raja: field we are giving our key. So, so, what I feel is we are not really operating our key, but giving it to stagehand, and.
305 00:24:58.560 ⇒ 00:24:59.679 Samuel Roberts: Exactly, yeah.
306 00:24:59.680 ⇒ 00:25:00.719 Mustafa Raja: is using that.
307 00:25:01.150 ⇒ 00:25:02.590 Samuel Roberts: I’m just trying to see…
308 00:25:03.430 ⇒ 00:25:08.690 Mustafa Raja: If StageHand gives us some observability, we can…
309 00:25:08.690 ⇒ 00:25:12.379 Samuel Roberts: Yeah, that’s what I’m hoping. As I’m looking at the docs right now, looks like you can get…
310 00:25:13.010 ⇒ 00:25:15.990 Mustafa Raja: Yeah, observe… yeah, yeah, observability.
311 00:25:15.990 ⇒ 00:25:18.289 Samuel Roberts: Final metrics, you can pull metrics from it.
312 00:25:18.440 ⇒ 00:25:24.419 Samuel Roberts: What is that for? What is that under? Performance tracking? That’s local.
313 00:25:27.790 ⇒ 00:25:29.590 Mustafa Raja: Yeah, LLM usage, no?
314 00:25:30.470 ⇒ 00:25:33.249 Samuel Roberts: Yeah, exactly, that’s exactly where I am. Yeah, yeah, you found it.
315 00:25:37.270 ⇒ 00:25:41.289 Mustafa Raja: Yeah, So I’ll have to set this up, right?
316 00:25:43.600 ⇒ 00:25:47.450 Samuel Roberts: Yeah, look at what, what part are you at? Configuration eval…
317 00:25:47.450 ⇒ 00:25:53.029 Mustafa Raja: Evaluations and… Okay. And I came over here from…
318 00:25:53.030 ⇒ 00:25:53.360 Samuel Roberts: Okay.
319 00:25:53.360 ⇒ 00:25:55.179 Mustafa Raja: If you use monitoring.
320 00:25:55.180 ⇒ 00:26:01.060 Samuel Roberts: Yeah, okay, I’m… I was on the real-time metrics and monitoring basic usage, but I guess, yeah, we probably want…
321 00:26:04.780 ⇒ 00:26:05.899 Samuel Roberts: There it is, okay.
322 00:26:07.070 ⇒ 00:26:08.340 Mustafa Raja: LLM usage.
323 00:26:08.340 ⇒ 00:26:10.139 Samuel Roberts: Yeah, I see it now, okay.
324 00:26:12.150 ⇒ 00:26:13.740 Samuel Roberts: Oh boy, where’s the…
325 00:26:14.220 ⇒ 00:26:19.050 Mustafa Raja: There it is, cost here. Is it really? Cost control, hmm.
326 00:26:21.280 ⇒ 00:26:23.030 Mustafa Raja: Oh, tomatoes.
327 00:26:24.220 ⇒ 00:26:27.839 Mustafa Raja: I guess we just need the costs.
328 00:26:28.350 ⇒ 00:26:30.380 Samuel Roberts: Yeah, you need to know the tofu Nice thing.
329 00:26:30.380 ⇒ 00:26:31.950 Mustafa Raja: is a little…
330 00:26:33.200 ⇒ 00:26:35.750 Samuel Roberts: Yeah, go back to the previous page with the observability.
331 00:26:38.700 ⇒ 00:26:40.290 Mustafa Raja: Sorry.
332 00:26:45.720 ⇒ 00:26:50.930 Samuel Roberts: And then, go to real-time metrics and Monitoring.
333 00:26:53.490 ⇒ 00:26:55.370 Samuel Roberts: And I think this…
334 00:26:56.600 ⇒ 00:26:57.490 Mustafa Raja: Oh, yeah.
335 00:26:57.860 ⇒ 00:27:00.309 Samuel Roberts: Yeah, I think that is probably the closest thing.
336 00:27:00.310 ⇒ 00:27:01.110 Mustafa Raja: Yeah, yeah, yeah.
337 00:27:01.110 ⇒ 00:27:06.500 Samuel Roberts: Because it looks like you can do prompt tokens and completion tokens, and then do a quick calculation there.
338 00:27:07.600 ⇒ 00:27:08.160 Mustafa Raja: Okay, good.
339 00:27:08.160 ⇒ 00:27:09.030 Samuel Roberts: Maybe try to act.
340 00:27:09.830 ⇒ 00:27:10.490 Samuel Roberts: Yeah.
341 00:27:10.770 ⇒ 00:27:17.610 Samuel Roberts: Yeah, yeah, yeah. Yeah, maybe, maybe I’ll run some random 10 websites? Yeah, exactly.
342 00:27:17.610 ⇒ 00:27:24.930 Mustafa Raja: With this code, and then average of that would be a better… Better idea, right?
343 00:27:25.270 ⇒ 00:27:26.310 Samuel Roberts: Yeah, I think so.
344 00:27:27.990 ⇒ 00:27:28.480 Mustafa Raja: Okay.
345 00:27:28.480 ⇒ 00:27:34.500 Samuel Roberts: I would go with that. Cool. Okay. Yeah, so try to get that together, and we’ll…
346 00:27:34.710 ⇒ 00:27:37.239 Samuel Roberts: See how bad it is and present it to them, I guess.
347 00:27:37.970 ⇒ 00:27:38.450 Mustafa Raja: Yeah.
348 00:27:38.660 ⇒ 00:27:42.460 Samuel Roberts: I just don’t know what their budget’s like, so I don’t want to, like, speak for them, you know?
349 00:27:45.130 ⇒ 00:27:50.169 Samuel Roberts: But yeah, get the information together is a good idea. Plus, this is gonna help. We need to start doing this more,
350 00:27:50.790 ⇒ 00:28:00.020 Samuel Roberts: You know, the other option here… Is… I was exploring, What is it, heli… Helicone?
351 00:28:08.050 ⇒ 00:28:10.610 Samuel Roberts: was it Pelicone, the gateway that.
352 00:28:10.610 ⇒ 00:28:11.080 Casie Aviles: Yeah.
353 00:28:11.080 ⇒ 00:28:21.469 Samuel Roberts: We were looking at, yeah. So that’s the other potential option for this sort of stuff, is a, like, gateway that can monitor all the calls that are going through an API key or something.
354 00:28:22.780 ⇒ 00:28:31.490 Samuel Roberts: But I think for this project, let’s just do it this way for now, but we may want to, in the future, look at that a little bit closer, if we really need to monitor, like, client usage of this stuff.
355 00:28:31.940 ⇒ 00:28:33.009 Mustafa Raja: Okay.
356 00:28:33.060 ⇒ 00:28:34.809 Samuel Roberts: But I think this is the way to do it for now.
357 00:28:35.770 ⇒ 00:28:36.420 Mustafa Raja: Okay.
358 00:28:37.330 ⇒ 00:28:37.950 Samuel Roberts: Great.
359 00:28:39.800 ⇒ 00:28:41.629 Samuel Roberts: What else? Because I gotta hop off.
360 00:28:41.630 ⇒ 00:28:42.100 Mustafa Raja: I shouldn’t.
361 00:28:42.100 ⇒ 00:28:43.010 Samuel Roberts: We’ll have to hop off.
362 00:28:43.010 ⇒ 00:28:43.550 Mustafa Raja: Oh, yeah, yeah.
363 00:28:43.550 ⇒ 00:28:44.160 Samuel Roberts: Yeah, with Mika.
364 00:28:44.160 ⇒ 00:28:46.720 Mustafa Raja: One more weird thing, if you want to talk about that.
365 00:28:46.720 ⇒ 00:28:48.650 Samuel Roberts: Yeah, yeah, real quick. Oh, yeah, real quick.
366 00:28:48.980 ⇒ 00:28:50.959 Mustafa Raja: Real quick, okay.
367 00:28:51.100 ⇒ 00:29:07.330 Mustafa Raja: Yeah, so all these rows got pulled, right? Yeah. And, I wanted to do a run on this new table that I have been crafting. I couldn’t last night because,
368 00:29:07.330 ⇒ 00:29:16.940 Mustafa Raja: I believe that our clay instance was occupied processing all of these, and it could not process that. Oh.
369 00:29:17.260 ⇒ 00:29:22.959 Mustafa Raja: And even trying to stop them was taking a lot of time. I woke
370 00:29:23.280 ⇒ 00:29:28.599 Mustafa Raja: morning, and it was still trying to stop. At the end, I just deleted those rows.
371 00:29:29.420 ⇒ 00:29:36.370 Mustafa Raja: So then our instance will be somewhat free, and I’ll be able to run these rows.
372 00:29:37.430 ⇒ 00:29:38.300 Samuel Roberts: Okay.
373 00:29:38.710 ⇒ 00:29:41.769 Mustafa Raja: This is something that comes in terms of scalability.
374 00:29:42.100 ⇒ 00:29:43.830 Samuel Roberts: Yeah, Clay may not be…
375 00:29:44.080 ⇒ 00:29:50.389 Mustafa Raja: Yeah, either that, or it ha- or it’s a problem with those specific columns.
376 00:29:51.100 ⇒ 00:29:53.999 Mustafa Raja: Those were lookup columns.
377 00:29:54.410 ⇒ 00:30:01.110 Mustafa Raja: Okay. So it’s either an issue with lookup, or issue with clay. I’m not sure about that right now.
378 00:30:01.440 ⇒ 00:30:04.679 Samuel Roberts: Alright, well, it’s good to… good to know. We’ll have to keep an eye on that, I think, more than anything.
379 00:30:04.680 ⇒ 00:30:06.760 Mustafa Raja: Yeah, yeah, I guess, we need to…
380 00:30:07.230 ⇒ 00:30:08.830 Samuel Roberts: Yes, we gotta run to the…
381 00:30:08.830 ⇒ 00:30:09.270 Mustafa Raja: Yeah.
382 00:30:09.270 ⇒ 00:30:10.120 Samuel Roberts: A good one. Okay.
383 00:30:10.510 ⇒ 00:30:11.810 Samuel Roberts: Alright, thank you guys.
384 00:30:12.870 ⇒ 00:30:13.390 Casie Aviles: Thank you.
385 00:30:13.390 ⇒ 00:30:14.670 Mustafa Raja: Thank you. Bye.