Meeting Title: US x BF | Standup Date: 2025-08-20 Meeting participants: Uttam Kumaran, Emily Giant, Demilade Agboola, Amber Lin


WEBVTT

1 00:00:48.590 00:00:49.850 Emily Giant: Hello!

2 00:00:50.440 00:00:51.550 Uttam Kumaran: Hello?

3 00:00:52.060 00:00:53.099 Emily Giant: How’s it going?

4 00:00:53.460 00:00:57.189 Uttam Kumaran: Good! Made, like, a lot of progress on, …

5 00:00:57.560 00:01:00.560 Uttam Kumaran: meta playing yesterday, so I’ll share some of that.

6 00:01:01.000 00:01:01.880 Emily Giant: Nice.

7 00:01:02.260 00:01:03.010 Uttam Kumaran: Yeah.

8 00:01:09.440 00:01:14.820 Emily Giant: Yeah, I saw in the channel that things were happening, and I was like, oh, cool, I like this.

9 00:01:14.820 00:01:16.809 Uttam Kumaran: We… we’re ha- we’re gonna have…

10 00:01:17.230 00:01:24.519 Uttam Kumaran: sort of observability over most of the events. I’m just, like, kind of figuring out a couple of things, but it should be…

11 00:01:25.310 00:01:28.959 Uttam Kumaran: A lot faster for us to spot issues, you know.

12 00:01:28.960 00:01:30.480 Emily Giant: Yeah, that’s awesome.

13 00:01:34.200 00:01:35.480 Amber Lin: Hello!

14 00:01:40.120 00:01:48.410 Amber Lin: Quickly go over… Tickets, and then we can talk about what we want to do for…

15 00:01:48.520 00:01:50.809 Amber Lin: Tomorrow’s meeting with the analysts.

16 00:01:51.150 00:01:51.540 Uttam Kumaran: book.

17 00:01:51.540 00:01:56.290 Amber Lin: Are all the… Tickets up today.

18 00:01:57.100 00:01:58.600 Uttam Kumaran: ….

19 00:01:59.560 00:02:01.539 Emily Giant: Let me check real quick on mine.

20 00:02:06.870 00:02:08.870 Uttam Kumaran: So, I kind of spent a lot of my…

21 00:02:09.229 00:02:18.460 Uttam Kumaran: day yesterday of finishing up, like, this Metaplane work. Just do a brief demo if we have a little bit of time at the end of this meeting.

22 00:02:18.740 00:02:23.400 Uttam Kumaran: We just have had these, like, job failures, and sort of, like.

23 00:02:23.560 00:02:28.730 Uttam Kumaran: you know, long-running jobs, and I… and I sort of just wanted to, like, put the…

24 00:02:30.460 00:02:36.490 Uttam Kumaran: let the brakes on it a little bit, so I have some stuff to share there, but overall, I feel pretty good.

25 00:02:36.820 00:02:39.600 Uttam Kumaran: Okay. And then, you know, ideally.

26 00:02:39.720 00:02:53.150 Uttam Kumaran: I’ve talked to the Metaplain team and said, like, hey, if we can get, like, a couple weeks to just, like, try it out and get familiar, and then… it’s roughly, like, 500 bucks a month, I think maybe a little bit cheaper.

27 00:02:53.390 00:02:54.649 Uttam Kumaran: Which…

28 00:02:55.100 00:03:06.869 Uttam Kumaran: I think it’s well worth it, because we’re… we’re gonna keep having issues before we fully transition stuff over, and even after. So, Emily, I’ll kind of show you what the layout is, and Emilada as well, and… and…

29 00:03:07.100 00:03:08.860 Uttam Kumaran: Yeah, you guys would be happy.

30 00:03:09.630 00:03:10.460 Emily Giant: Aw.

31 00:03:10.460 00:03:11.240 Amber Lin: Okay.

32 00:03:15.870 00:03:20.690 Amber Lin: Emily, anything here that can be… can be closed?

33 00:03:20.690 00:03:21.310 Emily Giant: Researching.

34 00:03:21.310 00:03:22.260 Amber Lin: status.

35 00:03:22.760 00:03:23.390 Emily Giant: Nope.

36 00:03:28.960 00:03:30.729 Uttam Kumaran: I moved one of mine.

37 00:03:41.170 00:03:42.350 Amber Lin: Whoa.

38 00:03:49.430 00:03:50.580 Amber Lin: Okay.

39 00:03:51.170 00:03:57.959 Demilade Agboola: From my end, I have… well, I’m about to merge the PR that I sent in for the split line items.

40 00:03:57.960 00:04:02.139 Amber Lin: So that can be marked as done, because Emily has reviewed it.

41 00:04:02.430 00:04:16.049 Demilade Agboola: And then… no, I’ve actually just marked as done. And then, so I’m kind of looking at the line items and fact line items, and I will probably create a…

42 00:04:16.649 00:04:24.650 Demilade Agboola: I’ll do them together, basically. I would push them together, because they’re quite similar models. So, they’re in progress right now.

43 00:04:25.280 00:04:32.119 Amber Lin: Okay, sounds good. … I think for the rest of the meeting, we can talk about…

44 00:04:32.410 00:04:42.740 Amber Lin: tomorrow’s meeting with the analysts, and then the metaplane, thing that we want to do. What do we want to talk about tomorrow when we meet with the analysts?

45 00:04:44.150 00:04:47.090 Emily Giant: Would Metaplane be something that…

46 00:04:47.260 00:04:51.290 Emily Giant: would be interesting to share? Like, I think as an analyst.

47 00:04:52.030 00:04:53.970 Uttam Kumaran: Yeah, I’m happy to share.

48 00:04:53.970 00:04:54.450 Emily Giant: Okay.

49 00:04:54.450 00:05:12.650 Uttam Kumaran: basically, I want to give people confidence that, like, yes, errors are going to happen, but our time to resolve them is going down, and roughly, we’re sort of building, like, a little bit of a, how do we track jobs succeeding job failure? So I can do a little bit on…

50 00:05:13.360 00:05:16.150 Uttam Kumaran: Like, observability and reliability.

51 00:05:16.330 00:05:17.010 Uttam Kumaran: like, maybe.

52 00:05:17.010 00:05:17.579 Emily Giant: Yeah, I….

53 00:05:17.580 00:05:18.120 Uttam Kumaran: your trust.

54 00:05:18.120 00:05:19.750 Emily Giant: I think that would be great, like…

55 00:05:19.980 00:05:25.510 Emily Giant: I think observability is a big miss right now, so….

56 00:05:25.510 00:05:25.950 Uttam Kumaran: Yeah.

57 00:05:25.950 00:05:34.450 Emily Giant: That would be great. And then, Demolade, if there’s any progress we can share on, like, the split line items thing, and how we plan to, like.

58 00:05:34.770 00:05:38.550 Emily Giant: resolve it. That would probably be an exciting

59 00:05:39.010 00:05:45.090 Emily Giant: topic, but that just depends on, like, where you’re at. We could maybe push that to next time, if that’s not…

60 00:05:45.190 00:05:46.170 Emily Giant: Ready.

61 00:05:46.570 00:05:55.410 Demilade Agboola: I mean, I could show the numbers, I… it’s not necessarily at the point where… The numbers are…

62 00:05:56.370 00:06:12.380 Demilade Agboola: I would say… I would say business stakeholder ready, like, it’s not… it’s more of like, oh, this model has been done, we’re aggregating… we have, like, all the sales and refunds, and we’re aggregating discounts and taxes and all of that, but…

63 00:06:12.590 00:06:17.529 Demilade Agboola: It doesn’t actually tell… it doesn’t actually tell a story yet, it’s just more of, like, right now.

64 00:06:18.300 00:06:20.990 Emily Giant: Yeah, maybe then next time, we do that.

65 00:06:23.750 00:06:24.840 Demilade Agboola: Well, fair enough.

66 00:06:25.610 00:06:34.160 Demilade Agboola: Because I know when… I don’t… were you in the call last, like, the past week, or… the call was 2 weeks ago, actually. Were you in that call?

67 00:06:35.860 00:06:37.610 Emily Giant: No, I was sick that day.

68 00:06:37.610 00:06:41.079 Demilade Agboola: Oh, okay. So, I… I think…

69 00:06:41.310 00:06:46.779 Demilade Agboola: I had some bits about, like, some of the numbers that we had in Looker, and I also.

70 00:06:46.780 00:06:47.130 Emily Giant: That’s amazing.

71 00:06:47.470 00:06:51.380 Demilade Agboola: like our technical design documents, and I talked about both parts.

72 00:06:51.800 00:06:57.240 Demilade Agboola: And… as you would expect, people were more responsive to the numbers that we had in Looker.

73 00:06:57.550 00:07:00.990 Demilade Agboola: Versus, like, the technical design documents, and I think it’s just…

74 00:07:01.990 00:07:07.380 Demilade Agboola: People gravitate to, you know, the numbers they can work with, or they can, you know.

75 00:07:07.380 00:07:07.970 Emily Giant: Yeah.

76 00:07:08.230 00:07:09.940 Demilade Agboola: It’s more easily done.

77 00:07:10.530 00:07:13.600 Demilade Agboola: Like, more technical stuff, and…

78 00:07:14.180 00:07:19.290 Demilade Agboola: I guess sometimes… well, obviously, we should brief them about what’s going on from a technical perspective.

79 00:07:19.650 00:07:34.370 Demilade Agboola: It would also be great to, like, go, hey, this is… I think, like, Metaplain would be great, because even though they might not have… they might have some questions, but, like, ultimately, it’s, like, data security, and, like, data, like, not security, but data, like, quality.

80 00:07:34.630 00:07:40.459 Demilade Agboola: And data, … so they have questions about that, and maybe they’ve had things affect them.

81 00:07:40.770 00:07:44.719 Demilade Agboola: In their place, like, in what they’ve been doing.

82 00:07:44.980 00:07:51.050 Demilade Agboola: But, like, strike line items, unless, unless, split line items, unless someone has had issues with that.

83 00:07:51.220 00:07:57.359 Demilade Agboola: They might not necessarily, like, have questions. It would just be like, okay, it’s coming, basically.

84 00:07:58.160 00:07:59.130 Demilade Agboola: Yeah.

85 00:08:03.630 00:08:11.669 Amber Lin: Okay, do we wanna… prepare anything for that, or are we just gonna demo Metaplane?

86 00:08:14.080 00:08:16.500 Demilade Agboola: Yeah, I think we could just demo Metaplane, …

87 00:08:16.710 00:08:21.660 Demilade Agboola: We can also just give an update, just generally, about what, like, what’s been going on in terms of revenue.

88 00:08:21.790 00:08:22.580 Demilade Agboola: Okay.

89 00:08:22.580 00:08:29.240 Uttam Kumaran: Yeah, Amber, are you typically doing slides for this, or what’s the typical… Or we just talk.

90 00:08:31.290 00:08:36.300 Demilade Agboola: sometimes slides, but, like, the last one, I kind of just talked and walked through stuff.

91 00:08:36.710 00:08:37.779 Demilade Agboola: I don’t like that.

92 00:08:41.120 00:08:45.210 Uttam Kumaran: Yeah, I mean, I could just share my screen, walk through Metaplane, …

93 00:08:45.900 00:08:55.189 Uttam Kumaran: And generally, it would be great to hear how people think about observability and stability, so I can ask them for any other things that we can improve on.

94 00:08:58.420 00:09:03.590 Emily Giant: Yeah, that sounds good. Like, keep it simple with what we can actually demo.

95 00:09:04.000 00:09:04.620 Uttam Kumaran: Okay.

96 00:09:12.230 00:09:18.090 Amber Lin: Checking tomorrow, let me just check, Uta, if you’re free at that meeting time.

97 00:09:18.980 00:09:20.280 Amber Lin: Okay.

98 00:09:21.010 00:09:28.640 Amber Lin: Yeah, it should be good tomorrow. That will be at your 10… 10.30.

99 00:09:32.470 00:09:33.030 Uttam Kumaran: Okay.

100 00:09:33.180 00:09:34.060 Amber Lin: Alright.

101 00:09:34.910 00:09:35.720 Amber Lin: Okay.

102 00:09:36.160 00:09:36.770 Emily Giant: Cool!

103 00:09:37.010 00:09:40.570 Uttam Kumaran: If I could just take 2 minutes and just walk through your Metaplane stuff?

104 00:09:40.840 00:09:41.420 Amber Lin: Yeah.

105 00:09:41.940 00:09:42.470 Emily Giant: Yeah.

106 00:09:43.320 00:09:48.110 Uttam Kumaran: Cool, so this is a long time coming, but I wanted to…

107 00:09:48.420 00:09:50.890 Uttam Kumaran: Kinda got a bunch of stuff done yesterday.

108 00:09:51.080 00:09:57.620 Uttam Kumaran: So, roughly, this is our Metaplane instance. On our side, we’ve connected

109 00:09:57.660 00:10:14.329 Uttam Kumaran: they kind of have this concept of data stacks. We’ve connected Redshift, dbt Cloud, and Looker. I would say focus for this is really on Redshift and dbt Cloud. So, a couple of things that, Metaplane does at a very basic level is there’s a concept of monitors.

110 00:10:14.440 00:10:17.789 Uttam Kumaran: So, a monitor is a…

111 00:10:17.970 00:10:23.879 Uttam Kumaran: sort of framework where you could track a metric related to an object. So, for example, you can track…

112 00:10:24.060 00:10:32.319 Uttam Kumaran: how long it’s been since a created at column has changed relative to the current time. You can look at

113 00:10:32.440 00:10:36.249 Uttam Kumaran: The row count for a table object.

114 00:10:36.360 00:10:42.829 Uttam Kumaran: And so what we’ve… what I’ve kind of done is I’ve gone ahead and set up several monitors across the stack.

115 00:10:43.020 00:10:54.990 Uttam Kumaran: So right now, I would say we have, really our core monitors here are on our analytics schema, as well as on our,

116 00:10:55.370 00:10:58.920 Uttam Kumaran: our dbt Brainforge schema, …

117 00:10:58.940 00:11:13.269 Uttam Kumaran: So, one, it’s on the… just the… what I… what I’ve done is I basically looked at the top 10% of tables that are queried from analytics, and we’ve added several monitors, freshness, row count. Similarly.

118 00:11:13.270 00:11:19.070 Uttam Kumaran: for our new work that’s in dbt Brainforge, I’ve also added, …

119 00:11:19.730 00:11:30.180 Uttam Kumaran: information there. I’ve also added monitors there, which, if I could get to… …

120 00:11:32.120 00:11:41.890 Uttam Kumaran: Yeah, so we also have monitors on several of these items. So if I go into this Redshift data stack, you can see that this is sort of the state of our current data stack right now.

121 00:11:42.070 00:11:43.020 Uttam Kumaran: …

122 00:11:43.190 00:11:50.000 Uttam Kumaran: we can see some helpful things, but I would say the most important here is just to start looking at these monitors. So, as you see, I’ve…

123 00:11:50.130 00:12:01.560 Uttam Kumaran: I’ve sort of set this up, and today it looks like there’s some errors, which… yeah, this is kind of, like, what I thought it was gonna be. I need to put a post hub to grant access to, …

124 00:12:01.970 00:12:07.660 Uttam Kumaran: the metaplane. But basically, if you look at the state of, let’s say, …

125 00:12:07.920 00:12:10.930 Uttam Kumaran: Where’s one that I was looking at yesterday?

126 00:12:13.610 00:12:19.249 Uttam Kumaran: Yeah, like… Let’s look at, like, Components XF.

127 00:12:19.530 00:12:24.519 Uttam Kumaran: As you can see here, this is a freshness on the created app column.

128 00:12:24.780 00:12:32.119 Uttam Kumaran: And so, what this is doing right now is it’s training. So you can see that every hour, it’s finding out what the value is.

129 00:12:32.270 00:12:35.789 Uttam Kumaran: And then it’s creating the bounds, right?

130 00:12:36.010 00:12:44.640 Uttam Kumaran: So, it looks like it’s 57, 58, 53, 30, 37. What it’s gonna arrive at is probably gonna be between

131 00:12:45.010 00:12:49.099 Uttam Kumaran: Like, it’s probably gonna end it at an hour or an hour and a half.

132 00:12:49.330 00:12:55.910 Uttam Kumaran: So, what that means is anytime this exceeds that, we’re gonna get an alert to Urban STEM’s data alerts.

133 00:12:56.100 00:13:03.059 Uttam Kumaran: So this is gonna take a little bit of time for, one, for it to sort of, like, understand what these are. Second.

134 00:13:03.160 00:13:06.780 Uttam Kumaran: We’re probably gonna be in a little bit of, like, heavy alerts for a sec.

135 00:13:07.120 00:13:12.600 Uttam Kumaran: Ideally, what we could start to do is say, are we okay with this SLA, or should we…

136 00:13:12.760 00:13:27.940 Uttam Kumaran: change the sensitivity. So one thing that we can do here is we can actually change the standard deviation, change the sensitivity, kind of improve that. But this is just one example of one column and one table, that’s being checked now every hour.

137 00:13:28.090 00:13:34.459 Uttam Kumaran: And so, my main goal for Redshift was get tracking on all of the existing XF tables.

138 00:13:34.640 00:13:43.109 Uttam Kumaran: And additionally, start implement tracking on our inventory marks, and then next up will be revenue mark.

139 00:13:43.370 00:14:00.610 Uttam Kumaran: Another concept in Metaplane is rules. So, if we actually start to invest in, like, tagging within dbt, like, we can tag everything from the new inventory mark, we can actually automatically monitor new tables that come in based on those tags.

140 00:14:00.640 00:14:10.699 Uttam Kumaran: So that’s also something that we’ll start to do. And row count… on tables, we have two types of monitors. We have column, and we have… we can have… also look at row counts.

141 00:14:10.850 00:14:14.249 Uttam Kumaran: Row counts is probably most important, is, like.

142 00:14:14.380 00:14:22.169 Uttam Kumaran: if, for example, if the cardinality breaks, and, like, we get a ton of duplicates, we want to be alerted. Created at may…

143 00:14:22.410 00:14:27.089 Uttam Kumaran: Maybe fine, but we want to say that, hey, this table jumped from a million to 10 million rows.

144 00:14:27.400 00:14:32.039 Uttam Kumaran: So, it’s a little bit of a work in progress, but I’ve kind of set up the baseline here.

145 00:14:32.340 00:14:35.159 Uttam Kumaran: Any questions on….

146 00:14:36.340 00:14:44.139 Emily Giant: Yeah, our… oh, not Redshift. No, I was gonna ask about our staging, which may… Be applicable, but, like.

147 00:14:44.140 00:14:44.840 Uttam Kumaran: Our staging it.

148 00:14:44.840 00:14:46.749 Emily Giant: environments, I’m curious, too.

149 00:14:46.750 00:14:57.820 Uttam Kumaran: Yeah, so let’s talk about dbt Cloud, and we can talk about… so on dbt Cloud stack, we can… we really can only look at job failure and job duration monitors.

150 00:14:57.940 00:15:13.770 Uttam Kumaran: And so, one thing we’re starting to look at here is, like, the job durations. These failed this morning, and the reason we didn’t get alerts is I just did not press a button here last night, so we should be getting alerts now. Right now, I’m having

151 00:15:13.990 00:15:22.000 Uttam Kumaran: Anything related to, … to dbt, get sent into that channel.

152 00:15:22.090 00:15:36.129 Uttam Kumaran: And so, for example, let’s look at, 60-minute off-peak core model refresh. This has been now running, for a little bit here, and so it’s gonna start to track what the model timings are.

153 00:15:36.180 00:15:43.440 Uttam Kumaran: Additionally, within one job, we can actually start to look at what is taking time.

154 00:15:43.980 00:15:46.460 Uttam Kumaran: Mmm, that’s cool. And this is, like…

155 00:15:46.620 00:15:52.599 Uttam Kumaran: It’s kind of a shame that dbt has not just done this. This would take them, like, 10 minutes to build, but…

156 00:15:52.740 00:15:53.860 Uttam Kumaran: Whatever.

157 00:15:54.260 00:16:00.320 Uttam Kumaran: … this is… this is, like, what I’m curious about. Like, I want to see, …

158 00:16:00.680 00:16:05.489 Uttam Kumaran: I forget what the name of this graph is. It’s not a Gantt chart, but it is some type of, like.

159 00:16:05.980 00:16:09.300 Uttam Kumaran: some type of, like, orchestration chart where I want….

160 00:16:09.650 00:16:16.909 Uttam Kumaran: dependencies in parallel. For example, we have 1, 2, 3, 4, 5… we have 8 threads running on this.

161 00:16:17.140 00:16:20.579 Uttam Kumaran: And you can see what… the key culprits are.

162 00:16:20.740 00:16:21.270 Uttam Kumaran: Okay.

163 00:16:21.270 00:16:21.680 Emily Giant: Yeah.

164 00:16:21.680 00:16:26.920 Uttam Kumaran: Now we have, like, what should we actually go attack in terms of incrementality? Easy.

165 00:16:27.160 00:16:29.259 Uttam Kumaran: Right? And so….

166 00:16:29.260 00:16:32.770 Emily Giant: Some of these are not even, like, real tables. That’s….

167 00:16:33.540 00:16:34.170 Uttam Kumaran: Yeah, I….

168 00:16:34.170 00:16:35.290 Emily Giant: wild.

169 00:16:35.290 00:16:38.700 Uttam Kumaran: So one thing that… one thing we can probably do is, like.

170 00:16:39.070 00:16:43.479 Uttam Kumaran: every week, maybe take an hour or something, and, like, we could just comb through and, like.

171 00:16:43.660 00:16:45.510 Uttam Kumaran: Crush through some of these, but….

172 00:16:45.510 00:16:45.910 Emily Giant: Yeah.

173 00:16:45.910 00:16:47.460 Uttam Kumaran: Overall, we have this…

174 00:16:47.740 00:16:56.650 Uttam Kumaran: track now, and… which is really, really great. Like, if we go to the every 30 minutes polyatomic inventory models, what we should see…

175 00:16:57.690 00:17:02.270 Uttam Kumaran: Is, one of these taking…

176 00:17:03.210 00:17:05.819 Uttam Kumaran: Well, I’ll be this one didn’t work, let’s see.

177 00:17:09.790 00:17:12.839 Uttam Kumaran: Yeah, I mean, right now, there’s this, …

178 00:17:13.089 00:17:15.309 Uttam Kumaran: There’s one table in this that is not…

179 00:17:16.010 00:17:25.659 Uttam Kumaran: like, ball that I need to look at today, but this is sort of the overview of, like, what we’re looking at in dbt, and so I have monitors now on duration. The rules I…

180 00:17:26.089 00:17:33.679 Uttam Kumaran: Here, though, is that anything that doesn’t match PR or doesn’t match Hawk, which is ad hoc.

181 00:17:34.130 00:17:34.520 Emily Giant: Okay.

182 00:17:34.520 00:17:41.550 Uttam Kumaran: honored. … we have a pretty good naming convention now, we have scheduled versus… you know, app.

183 00:17:42.020 00:17:42.740 Uttam Kumaran: Bye.

184 00:17:42.740 00:17:46.240 Emily Giant: Yeah. That way, any new jobs that get created get tracked here.

185 00:17:47.530 00:17:51.130 Emily Giant: That’s great. For the stakeholders, I would just make sure to, like.

186 00:17:51.560 00:17:56.190 Emily Giant: clarify what DBT is, what Redshift is, and how that affects.

187 00:17:56.190 00:18:05.580 Uttam Kumaran: Looker, because they so much see their work as in Looker. Okay. That, yeah, they’ll get it. It’s not gonna be confusing, as long as they understand, like.

188 00:18:05.580 00:18:09.120 Emily Giant: That… those are what… fuels looker.

189 00:18:10.280 00:18:15.280 Uttam Kumaran: The other thing, you know, that we can start adding is dashboard monitors.

190 00:18:16.460 00:18:17.480 Uttam Kumaran: break…

191 00:18:17.670 00:18:26.439 Uttam Kumaran: all they have is sort of, like, total view monitors right now, which is not really that helpful. I think if we do a look…

192 00:18:27.540 00:18:33.640 Uttam Kumaran: Yeah, I’m not sure how great the Looker monitors are, …

193 00:18:33.820 00:18:37.789 Uttam Kumaran: It looks like mainly it’s just, like, tracking views and stuff.

194 00:18:38.180 00:18:46.159 Amber Lin: Hi there! I have another stand up. Would you guys mind using a different meeting room, or maybe using a different Zoom link?

195 00:18:46.160 00:18:49.910 Uttam Kumaran: Yeah, let me, let me just… I’ll just, do a huddle in the, in the channel, maybe.

196 00:18:49.910 00:18:50.850 Amber Lin: Okay, awesome.

197 00:18:50.850 00:18:52.010 Uttam Kumaran: Yeah, that’s fine.

198 00:18:52.010 00:18:52.410 Amber Lin: Thanks.

199 00:18:52.410 00:18:52.760 Demilade Agboola: Oh, no.

200 00:18:52.760 00:18:53.410 Emily Giant: Alright.

201 00:18:53.410 00:18:54.390 Amber Lin: Bye. Bye.