Meeting Title: Magic Spoon SPINS sync Date: 2026-01-30 Meeting participants: Ashwini Sharma, Uttam Kumaran, Demilade Agboola, Awaish Kumar


WEBVTT

1 00:00:34.130 00:00:34.990 Ashwini Sharma: Hey.

2 00:00:36.520 00:00:37.460 Ashwini Sharma: Hello.

3 00:00:38.830 00:00:39.580 Uttam Kumaran: Can you hear me?

4 00:00:40.520 00:00:41.770 Ashwini Sharma: I can now, yeah.

5 00:00:51.550 00:00:53.119 Demilade Agboola: Hello, how’s everyone doing?

6 00:00:53.470 00:00:54.509 Uttam Kumaran: Hey, good.

7 00:00:54.830 00:00:55.739 Ashwini Sharma: Good, good.

8 00:01:03.670 00:01:06.759 Demilade Agboola: So we have, like, 2 hours to the meeting.

9 00:01:07.340 00:01:14.130 Demilade Agboola: Do we have any things we want to, like, flag, or, you know, just ensure we’re in a good spot?

10 00:01:17.790 00:01:27.060 Ashwini Sharma: No, I… So, like, I had updated that sheet, which Utah had sent out,

11 00:01:27.590 00:01:38.979 Ashwini Sharma: So, I compared it with the aggregated data that is returned from the API. There are still some mismatches, and those mismatches seems like it’s coming from the API itself.

12 00:01:39.390 00:01:50.020 Ashwini Sharma: When I do a bulk extract, it sends me lesser data… oh, sorry, when I do a weekly extract, it sends me lesser data compared to when I do an aggregated extract.

13 00:01:51.960 00:01:52.540 Demilade Agboola: Okay.

14 00:01:52.940 00:01:58.809 Ashwini Sharma: Yeah, so that’s the thing we’ll have to investigate further, and then talk with Magic Spin’s people, and then figure it out.

15 00:01:59.070 00:02:03.180 Ashwini Sharma: But other than that, it matches very closely to,

16 00:02:03.430 00:02:07.480 Ashwini Sharma: do what’s there in the aggregated data from API.

17 00:02:07.650 00:02:15.130 Ashwini Sharma: I haven’t got a chance to look into the… they haven’t loaded, in fact, right, the platform data, so I cannot say what is there.

18 00:02:15.710 00:02:19.150 Ashwini Sharma: But, now we have, now it is in a pretty good shape.

19 00:02:20.380 00:02:25.520 Demilade Agboola: Okay, so for now, would you say our recommendation is that we should use the…

20 00:02:25.870 00:02:29.630 Demilade Agboola: We should aggregate where we can before hitting the warehouse.

21 00:02:29.800 00:02:30.780 Demilade Agboola: Or…

22 00:02:31.010 00:02:38.089 Demilade Agboola: like, I want us to be able to tell them, like, hey, for now, if we really need to use the data, let’s aggregate.

23 00:02:38.450 00:02:46.269 Demilade Agboola: from the API, rather than try to get the, like, get it granular and then aggregate. Would that be the… Oh, aggregating right now?

24 00:02:46.270 00:02:49.559 Ashwini Sharma: Yeah, aggregating from the API would be,

25 00:02:50.370 00:02:54.750 Ashwini Sharma: you know, too many API calls, right? It won’t, return us.

26 00:02:55.250 00:02:59.350 Ashwini Sharma: The entire data. We will run out of rate limits before the pipeline finishes.

27 00:03:00.150 00:03:07.700 Ashwini Sharma: Okay. What happens is, like, every month when they release the data, right, they update the prior records.

28 00:03:08.740 00:03:09.430 Demilade Agboola: Okay.

29 00:03:09.430 00:03:11.960 Ashwini Sharma: Okay, up to 26 weeks back.

30 00:03:12.430 00:03:17.019 Ashwini Sharma: So, if we get aggregated data, Right? From the API,

31 00:03:17.630 00:03:21.989 Ashwini Sharma: Then, what happens is, like, for every week, you have to aggregate it.

32 00:03:22.640 00:03:31.029 Ashwini Sharma: Right? Not just the ending week. We are looking at the ending week because that forms a baseline for us to do a comparison.

33 00:03:31.150 00:03:41.240 Ashwini Sharma: But think about it, like, who is going to use this data, right? It’s the customer who’s going to use it? And what are they going to use it for? Do some trend analysis, right? So, for example, they might.

34 00:03:41.240 00:03:41.640 Demilade Agboola: Okay.

35 00:03:41.640 00:03:43.250 Ashwini Sharma: Say, okay, what is the…

36 00:03:43.470 00:03:49.600 Ashwini Sharma: You know, what the data look like for a 4-week trend on week ending, let’s say.

37 00:03:50.420 00:03:53.400 Ashwini Sharma: December 31st, or something like that, right?

38 00:03:54.700 00:03:59.370 Ashwini Sharma: So, which means that we’ll have to, you know, our models should be able to clearly

39 00:03:59.780 00:04:01.500 Ashwini Sharma: And visualize that, right?

40 00:04:06.480 00:04:07.960 Demilade Agboola: Okay.

41 00:04:07.960 00:04:13.659 Ashwini Sharma: It’s easier to get all this information if we have one week of data, right? If we have data for every one week.

42 00:04:13.900 00:04:17.609 Ashwini Sharma: And we can always recreate that information, but if we…

43 00:04:17.959 00:04:22.350 Ashwini Sharma: Think about extracting it for aggregated levels, for every week.

44 00:04:22.460 00:04:24.170 Demilade Agboola: And that becomes, like…

45 00:04:24.300 00:04:28.790 Ashwini Sharma: you know, several thousands API calls, and that… that will definitely…

46 00:04:29.100 00:04:32.150 Ashwini Sharma: Obviously it’s calling into a hurdle, yeah.

47 00:04:32.940 00:04:33.990 Demilade Agboola: Okay, so…

48 00:04:34.220 00:04:34.740 Ashwini Sharma: Yeah.

49 00:04:35.560 00:04:39.610 Demilade Agboola: Ashwini, so I think this is what I want you to, like, just think of recommendation.

50 00:04:39.840 00:04:46.720 Demilade Agboola: The customer needs to know, or, like, the client’s magic spon needs to know, hey, we want to get this data out to the…

51 00:04:47.010 00:04:49.700 Demilade Agboola: To our clients, right?

52 00:04:49.800 00:04:52.460 Demilade Agboola: Do we need to be able to say, hey.

53 00:04:52.610 00:04:55.649 Demilade Agboola: Right now, we’ll suggest we’re going to work with spins.

54 00:04:55.840 00:05:00.340 Demilade Agboola: and wait for them to fix whatever issues we can see. Or…

55 00:05:00.500 00:05:07.259 Demilade Agboola: can we find a workaround? Which is why I asked you if we can aggregate directly from the API and use that in their reports.

56 00:05:07.490 00:05:14.500 Demilade Agboola: Because the idea is, how soon can we ship this out to them so that they can see it in their mouths and have it for them to use?

57 00:05:14.810 00:05:18.280 Ashwini Sharma: See, the data is already there. You can already use it for math.

58 00:05:19.590 00:05:23.700 Demilade Agboola: Yes, you’re still not… but you’re saying that there are things that are being dropped.

59 00:05:24.020 00:05:26.449 Ashwini Sharma: Yeah, some of the, some records have been.

60 00:05:26.450 00:05:27.200 Demilade Agboola: I understand.

61 00:05:27.200 00:05:30.020 Uttam Kumaran: But Ashwini, can you please, like, write it down?

62 00:05:30.760 00:05:33.189 Ashwini Sharma: Yeah, I’ll create a document.

63 00:05:33.650 00:05:43.809 Demilade Agboola: Yeah, but we can’t ship data where things are being dropped, because then they can’t trust the data to give their client. So we need to be able to know, hey, is this the plan for us to use

64 00:05:44.050 00:05:49.260 Demilade Agboola: And, like, ideally, if we use the aggregated API, it would work.

65 00:05:49.600 00:05:52.310 Demilade Agboola: But then, we run out of calls.

66 00:05:52.680 00:05:58.999 Demilade Agboola: So, we should probably wait for the Spins APM to connect with Spins, and they fix the problem.

67 00:05:59.710 00:06:06.069 Demilade Agboola: So that we all have a clear timeline of what’s going on, and no one has to, like, wonder what’s… what’s happening.

68 00:06:08.490 00:06:26.810 Demilade Agboola: So I just want us to have, like, that very clear idea of, hey, we can get you the data right now through API aggregation, but it will take too… it will take too many API calls. So what we will do instead is we will wait for Spins to fix the API, and then we can use raw data and aggregate it in dbt.

69 00:06:26.900 00:06:29.409 Demilade Agboola: For the same things, for the same numbers you want.

70 00:06:30.600 00:06:31.750 Ashwini Sharma: Yeah, we can do that.

71 00:06:32.330 00:06:40.410 Demilade Agboola: Okay. Yeah, so I just want us to document that and send it to them. So, like, when we’re sending, like, end-of-week reports to them, we can make it clear what our recommendations.

72 00:06:40.410 00:06:46.809 Uttam Kumaran: But how long is that gonna take, like, even putting that dock together? Like, can we just do that now, and, like, can we have a dock? Because…

73 00:06:47.100 00:06:52.679 Uttam Kumaran: I’m telling you, like, if we don’t get this out today, they’re gonna probably cancel the contract.

74 00:06:52.970 00:06:56.820 Uttam Kumaran: So, Ashwini, if you can just take everything you said and throw it into a dock.

75 00:06:57.090 00:07:00.370 Uttam Kumaran: And if you can collaborate with Awash to create that document.

76 00:07:00.650 00:07:06.259 Uttam Kumaran: It’s something we can review, and I can pass it to the team, so I can continue to, you know, translate this here.

77 00:07:06.870 00:07:10.840 Ashwini Sharma: Okay, I’m creating a document highlighting the same thing.

78 00:07:11.500 00:07:22.500 Uttam Kumaran: Okay, and Awash, do you mind collaborating? Like, ideally, we just want to show exactly the situations where we’re running out of API calls, or, like, the rows are getting dropped.

79 00:07:22.700 00:07:26.790 Uttam Kumaran: Ideally, the code that’s being executed to do so, and then…

80 00:07:27.120 00:07:29.370 Uttam Kumaran: What our sort of recommendation is.

81 00:07:29.560 00:07:32.490 Uttam Kumaran: If we can just have that in a clean one-pager.

82 00:07:33.050 00:07:35.899 Uttam Kumaran: That’s it, like, that’s all we need here.

83 00:07:36.720 00:07:38.580 Awaish Kumar: Yeah, I can collaborate.

84 00:07:39.560 00:07:40.110 Uttam Kumaran: Okay.

85 00:07:41.720 00:07:42.910 Uttam Kumaran: Okay, great, then I’m gonna…

86 00:07:43.250 00:07:44.700 Awaish Kumar: Yeah, sorry.

87 00:07:46.410 00:07:47.070 Uttam Kumaran: Sorry?

88 00:07:47.700 00:07:50.940 Awaish Kumar: I mean, we are only talking about Spin’s API, right?

89 00:07:51.230 00:07:52.259 Uttam Kumaran: Yeah, that’s correct.

90 00:08:02.140 00:08:06.759 Awaish Kumar: Okay, then… do you… do you mind, like, we can also discuss CTA?

91 00:08:08.050 00:08:17.669 Uttam Kumaran: Yeah, I mean, I… this is gonna be more of the focus, so maybe let’s just spend, like, 2 minutes on CTA, and then I have to drop to go prep for that call.

92 00:08:18.230 00:08:25.279 Uttam Kumaran: So… I… I’m just gonna go review the tickets that Ashwini created. Is there any other discussion points there?

93 00:08:25.830 00:08:29.239 Uttam Kumaran: I assume, like, we haven’t made much progress.

94 00:08:30.720 00:08:33.530 Awaish Kumar: I just wanted to assign tickets to us so we can…

95 00:08:33.539 00:08:37.159 Uttam Kumaran: Okay, let’s go ahead and… let’s go ahead and just do that now, then.

96 00:08:37.479 00:08:39.629 Uttam Kumaran: While we have a second here.

97 00:08:45.569 00:08:46.899 Uttam Kumaran: Oh yeah, you go ahead.

98 00:08:47.690 00:08:50.789 Ashwini Sharma: Now, go to next cycle. I just moved it to next.

99 00:08:53.760 00:08:54.330 Uttam Kumaran: Okay.

100 00:08:54.610 00:08:59.809 Uttam Kumaran: So, do these have, like, any notes in them?

101 00:09:00.080 00:09:08.400 Ashwini Sharma: No, we’ve… I’ve shared one document, right? A PDF document, so we’ll have to see that, and then go into the database, and then get.

102 00:09:11.310 00:09:11.880 Uttam Kumaran: Can you open that?

103 00:09:12.630 00:09:14.309 Uttam Kumaran: Yeah, so maybe…

104 00:09:14.310 00:09:18.599 Ashwini Sharma: Can you open that PDF document that I shared? It’s in the Slack channel.

105 00:09:23.140 00:09:26.730 Awaish Kumar: Send the… Transitute?

106 00:09:27.310 00:09:32.220 Ashwini Sharma: No, it’s a member engagement report. Client CTA, yeah. Yes.

107 00:09:39.280 00:09:45.310 Uttam Kumaran: So there’s, like, engagement, committee, participation… so you’re saying you broke it down into each of these sections?

108 00:09:45.310 00:09:46.389 Ashwini Sharma: Yeah, yeah.

109 00:09:47.490 00:09:50.540 Uttam Kumaran: And do you have access to this, or they didn’t give you access to this?

110 00:09:50.740 00:09:51.800 Ashwini Sharma: Access to what?

111 00:09:52.130 00:09:53.080 Uttam Kumaran: the report.

112 00:09:54.880 00:09:57.719 Ashwini Sharma: No, I just have a snapshot of that report, right?

113 00:09:57.720 00:10:01.830 Uttam Kumaran: For example, if you look at events attendance, looks like there’s more columns, right?

114 00:10:02.170 00:10:04.650 Uttam Kumaran: So how do we know all the columns for each?

115 00:10:04.980 00:10:10.860 Ashwini Sharma: One second, let me open that report. Which one? Events Attendance, attendance, this one, right?

116 00:10:11.370 00:10:15.170 Uttam Kumaran: First name, last name, title, event title, but it looks like it keeps going.

117 00:10:15.370 00:10:16.749 Ashwini Sharma: Hold on a second.

118 00:10:16.880 00:10:18.870 Ashwini Sharma: Where are you seeing that? Sorry.

119 00:10:19.870 00:10:21.229 Uttam Kumaran: If you open the PDF, and then.

120 00:10:21.230 00:10:21.550 Ashwini Sharma: Yeah.

121 00:10:21.550 00:10:24.820 Uttam Kumaran: Go to the bottom, go to Events Attendance.

122 00:10:24.990 00:10:29.419 Ashwini Sharma: Oh, this is… this is just a Power BI report. She took a snapshot of that, and then.

123 00:10:29.420 00:10:35.550 Uttam Kumaran: I know, I know, I know, but, like, let’s open the ticket for… open the ticket for data model for events attendance, Awash.

124 00:10:37.950 00:10:38.629 Awaish Kumar: This one…

125 00:10:38.870 00:10:40.230 Uttam Kumaran: There’s nothing in here.

126 00:10:40.380 00:10:45.680 Uttam Kumaran: Right, so then if I was to take this ticket, I’m gonna go to this sheet, I’m gonna say, cool, I see

127 00:10:46.490 00:10:50.349 Uttam Kumaran: Four columns, first name, last name, title, event title.

128 00:10:51.100 00:10:54.230 Uttam Kumaran: It looks like there’s several more columns, right?

129 00:10:54.560 00:10:56.250 Uttam Kumaran: Do we know what those columns are?

130 00:10:56.250 00:10:58.079 Ashwini Sharma: Title, session title, one more.

131 00:10:58.460 00:10:59.699 Uttam Kumaran: Is there anything else?

132 00:10:59.700 00:11:03.330 Ashwini Sharma: No, that’s it. Take a look at the right side report, right?

133 00:11:03.330 00:11:08.659 Uttam Kumaran: I know, but are you sure, or are you just… I’m looking at the same thing, but are you sure? Like, do you know this for fact?

134 00:11:08.860 00:11:11.210 Ashwini Sharma: This is for… no, I don’t have it as a fact.

135 00:11:11.210 00:11:11.840 Uttam Kumaran: Okay, okay.

136 00:11:11.840 00:11:13.559 Ashwini Sharma: 2024 and 2025.

137 00:11:14.060 00:11:18.640 Uttam Kumaran: No, no, no, I know, but, like, this is what I’m saying, like, yes, I’m seeing the same thing you’re saying, but, like.

138 00:11:18.640 00:11:19.110 Ashwini Sharma: Yup.

139 00:11:19.110 00:11:20.930 Uttam Kumaran: We don’t have the requirements.

140 00:11:21.190 00:11:26.380 Uttam Kumaran: And we need to ask them, like, hey, are… we need to be able to confirm the requirements before building.

141 00:11:26.620 00:11:29.649 Uttam Kumaran: Right? So, I think, Awish, how about, like.

142 00:11:30.040 00:11:37.579 Uttam Kumaran: do you have any sense, Ashwini, of, like, difficulty on any of these? Like, on terms of all the tickets? Like, what’s more difficult than the other?

143 00:11:37.820 00:11:42.960 Ashwini Sharma: The entity resolution is there, right, in most of them, which means, like.

144 00:11:43.540 00:11:50.960 Ashwini Sharma: You know, we’ll have to identify what is the record, for whom that record Is created, right?

145 00:11:51.160 00:11:59.109 Ashwini Sharma: I mean, any record that we see in a main table, right, we’ll have to figure that out, how it matches to an individual or to an organization.

146 00:11:59.640 00:12:04.309 Ashwini Sharma: Now, that could be directly through Impexium org ID, or it could be through…

147 00:12:04.560 00:12:09.140 Ashwini Sharma: An email, or a, you know, domain address, or alias, and things like that.

148 00:12:10.780 00:12:11.540 Awaish Kumar: Okay.

149 00:12:11.940 00:12:15.869 Awaish Kumar: So, like, what’s… So we are just…

150 00:12:15.870 00:12:19.800 Uttam Kumaran: She said, yeah, she… she just wants to be one of this whole thing.

151 00:12:19.920 00:12:24.739 Uttam Kumaran: I think it’s not clear to me what part of this is more difficult than the other.

152 00:12:24.910 00:12:29.760 Uttam Kumaran: I would prefer for me to be able to take whatever the toughest part is.

153 00:12:30.010 00:12:32.770 Uttam Kumaran: So that I could just learn a little bit about the data model.

154 00:12:32.960 00:12:36.730 Uttam Kumaran: Like, If it’s not clear, then…

155 00:12:37.150 00:12:42.769 Uttam Kumaran: I’ll just try to take, like, the top part, right? So, if you talk… start… is there an account summary?

156 00:12:44.450 00:12:44.810 Ashwini Sharma: In the morning.

157 00:12:45.150 00:12:48.320 Ashwini Sharma: part would be the CES engagement, right?

158 00:12:48.320 00:12:48.840 Uttam Kumaran: Okay.

159 00:12:50.360 00:12:54.669 Ashwini Sharma: There will be the core, entity resolution involved.

160 00:12:57.530 00:13:06.729 Uttam Kumaran: Okay, so can you… so maybe, Awash, you can take the media opportunities… And the other engagements?

161 00:13:09.600 00:13:13.380 Uttam Kumaran: matchmaker speakers.

162 00:13:15.900 00:13:18.039 Uttam Kumaran: And then event attendance.

163 00:13:19.460 00:13:22.330 Uttam Kumaran: Let me take the research downloads.

164 00:13:22.440 00:13:33.700 Uttam Kumaran: committee participation… Yeah, and research downloads… And then the engagement…

165 00:13:34.640 00:13:35.040 Awaish Kumar: We have…

166 00:13:35.040 00:13:41.850 Uttam Kumaran: Data model for CES standards…

167 00:13:41.980 00:13:43.910 Uttam Kumaran: Which one is this related to?

168 00:13:43.910 00:13:44.769 Ashwini Sharma: Alerts and DLCs.

169 00:13:44.770 00:13:47.750 Uttam Kumaran: Oh, okay, GLA, yeah, I can take this one also.

170 00:13:48.600 00:13:50.360 Uttam Kumaran: And then,

171 00:13:53.600 00:13:56.539 Uttam Kumaran: Registration and attendees, I’ll take that one too.

172 00:13:58.070 00:14:01.500 Uttam Kumaran: And then, what is the, EB bill?

173 00:14:03.720 00:14:07.790 Ashwini Sharma: There are also some amount of, INTD resolution is involved.

174 00:14:07.790 00:14:13.109 Uttam Kumaran: Oh, okay. Alright, I can take the EV bill, and then maybe I can assign the rest to Ashwini, that’s fine.

175 00:14:14.000 00:14:14.470 Awaish Kumar: Okay.

176 00:14:14.470 00:14:14.980 Ashwini Sharma: Correct.

177 00:14:16.370 00:14:29.470 Uttam Kumaran: But, yeah, let’s try to… if you… so I’m gonna go drop to just to try to work on this a little bit, and then I’ll prepare for the CTA meeting. Yeah, if you guys can stay on and wrap up Magic Spoon, our call is in about, again, hour and a half.

178 00:14:29.820 00:14:44.120 Uttam Kumaran: I would like to get something over to them, even just a one-pager report on, like, what help we need from Spins API. Today, I’m trying to drive them into two decisions. One, like, saying, okay, we have a clear path to resolving Spins.

179 00:14:44.150 00:14:52.149 Uttam Kumaran: And then second, Demi, like, you let me know if you feel comfortable at any point in the next hour and a half. I guess I’m gonna be kind of booked for the next hour, but…

180 00:14:52.210 00:14:59.279 Uttam Kumaran: I can review any PRs. A good win for us that could, like, kind of buy us some time is to get the model out, you know?

181 00:15:01.230 00:15:03.060 Demilade Agboola: Yeah, but, like, the…

182 00:15:03.690 00:15:08.149 Demilade Agboola: I’m almost done with the model. We will just basically need to run it with JT.

183 00:15:10.220 00:15:13.340 Uttam Kumaran: Like, I would like to… we could just take time in the call to do that.

184 00:15:14.280 00:15:30.850 Demilade Agboola: Yeah, yeah, sure, that’s not a problem. But, like, ultimately, I, like, would show what we’ve done. I have the… I have done the things, loaded the CSVs in, and I’m just, like, trying to test and be sure that the numbers are, like, they make a lot of sense.

185 00:15:30.910 00:15:35.459 Demilade Agboola: And the filters make a lot of sense, but that’s… that’s really just it. Finishing touches to everything.

186 00:15:37.070 00:15:37.640 Uttam Kumaran: Okay.

187 00:15:38.610 00:15:39.190 Demilade Agboola: Okay.

188 00:15:39.840 00:15:40.680 Uttam Kumaran: Okay, great.

189 00:15:42.670 00:15:44.810 Uttam Kumaran: Okay, thank you. Thank you, guys.

190 00:15:49.460 00:15:51.009 Awaish Kumar: Actually, do you realize that?

191 00:15:51.500 00:15:52.820 Ashwini Sharma: Yeah, I’m good.

192 00:15:53.330 00:15:54.170 Awaish Kumar: Yeah, that’s…

193 00:15:54.680 00:16:01.060 Ashwini Sharma: Okay, what do we want to do with the magic spawn? When’s the next meeting here? Hold on a second. Okay, in 15 minutes.

194 00:16:01.310 00:16:03.730 Awaish Kumar: You do have an argument for that, right?

195 00:16:04.620 00:16:06.700 Ashwini Sharma: I’m creating one document for that.

196 00:16:06.700 00:16:08.240 Awaish Kumar: No, I already… I think I…

197 00:16:08.770 00:16:13.230 Awaish Kumar: There was one document I wanted to… Have you…

198 00:16:13.400 00:16:20.339 Awaish Kumar: Document everything that… that you tried, and, like, this… Oh, mad what Stanley.

199 00:16:21.150 00:16:27.209 Ashwini Sharma: No, this is not what we are doing over there, right? It’s not about the pipeline, it’s about the API right now.

200 00:16:27.730 00:16:30.130 Ashwini Sharma: So, let’s put that information over here.

201 00:16:30.130 00:16:34.079 Awaish Kumar: Did Spins API work? No, sir. This was… or Spins API, right?

202 00:16:34.840 00:16:43.769 Ashwini Sharma: This is… this document that Otam talked about is for the Spence API, right? I mean, what we have done right now. The other one was… what… the other document? Yeah.

203 00:16:43.770 00:16:51.289 Awaish Kumar: It was also from Spins API, because you mentioned that there’s a lot of data, and that is why Spins API is failing to

204 00:16:51.490 00:16:53.269 Awaish Kumar: We are fully run the pipeline.

205 00:16:54.430 00:17:05.260 Ashwini Sharma: Yes. Here, there are… there are two things over here, right? Right now, we are talking about QAing thing, right? And what prevents us from QAing this thing, right?

206 00:17:05.349 00:17:22.700 Ashwini Sharma: Correctly. So that is what I want to highlight in this document. The other document was about the pipeline architecture, right? What I’ve done in the pipeline, what all different things that I’ve tried to make the pipeline faster, or run it properly, right? Those were the things. I don’t want to mix them together.

207 00:17:23.609 00:17:29.189 Awaish Kumar: No, but, so, you are saying now we have the data from this specific… Yes.

208 00:17:29.570 00:17:29.970 Ashwini Sharma: Yes.

209 00:17:29.970 00:17:35.189 Awaish Kumar: And… but we are still… We’re not able to QA, because we have some

210 00:17:35.660 00:17:40.029 Awaish Kumar: issues, or whatever, from… either from Swing’s API or wherever.

211 00:17:40.290 00:17:41.680 Awaish Kumar: Is that the case?

212 00:17:41.680 00:17:49.140 Ashwini Sharma: Okay, let me try to explain it, what is the case, right? So right now, we have data, right? We have data for every week, okay?

213 00:17:50.510 00:17:51.170 Ashwini Sharma: Okay.

214 00:17:51.660 00:18:06.149 Ashwini Sharma: And that is the, you know, least grain of data, right? I cannot, you know, slice that data down further, okay? That is the smallest grain of the data available. So, for example, like, yeah.

215 00:18:06.700 00:18:08.620 Awaish Kumar: Weekly is a grain, right?

216 00:18:08.770 00:18:10.269 Ashwini Sharma: Weekly data is the grain.

217 00:18:10.630 00:18:11.320 Awaish Kumar: Yeah.

218 00:18:11.320 00:18:17.009 Ashwini Sharma: Okay? Now, what we… what they want to analyze is, they want to analyze a 4-week report.

219 00:18:18.310 00:18:19.100 Awaish Kumar: Okay.

220 00:18:19.100 00:18:21.219 Ashwini Sharma: 5th, 12th week.

221 00:18:21.930 00:18:24.480 Ashwini Sharma: 24 weeks, and 52 weeks.

222 00:18:25.110 00:18:25.670 Awaish Kumar: Okay.

223 00:18:26.350 00:18:29.270 Ashwini Sharma: Okay? And this is from a certain date.

224 00:18:32.920 00:18:40.779 Ashwini Sharma: Alright? So, considering that the last date for which the data is available is 12-28-2025,

225 00:18:43.010 00:18:49.200 Ashwini Sharma: Right? If we consider this as a baseline, Right? 4-week report means that

226 00:18:52.820 00:18:56.680 Ashwini Sharma: means, like, glove.

227 00:18:56.840 00:19:01.500 Ashwini Sharma: 28, 2025, till… Four weeks back.

228 00:19:07.340 00:19:12.770 Ashwini Sharma: Alright? And similarly, 12-week report means… Since this week.

229 00:19:13.090 00:19:15.479 Ashwini Sharma: We go back in Perfix, and so on.

230 00:19:18.140 00:19:18.990 Ashwini Sharma: Okay.

231 00:19:21.020 00:19:22.620 Ashwini Sharma: Till this part, it’s clear, right?

232 00:19:24.180 00:19:25.729 Ashwini Sharma: This is what they want to see.

233 00:19:26.740 00:19:36.460 Ashwini Sharma: Okay? Now, let’s, how we get the data from Spins? I sent a request Spins, right?

234 00:19:37.060 00:19:40.759 Ashwini Sharma: Give me data for… One week.

235 00:19:42.550 00:19:46.670 Ashwini Sharma: Starting at… 1228.

236 00:19:48.480 00:19:53.900 Ashwini Sharma: Going back… Up to 52 weeks.

237 00:19:56.480 00:19:57.330 Ashwini Sharma: Alright.

238 00:19:58.020 00:20:01.090 Ashwini Sharma: this… this is what I can specify in one call.

239 00:20:01.580 00:20:02.440 Ashwini Sharma: Okay.

240 00:20:02.700 00:20:08.270 Ashwini Sharma: Or, what I can do is, I can also do… Request to spin.

241 00:20:09.040 00:20:12.009 Ashwini Sharma: Give me data for 4 weeks.

242 00:20:12.590 00:20:15.900 Ashwini Sharma: Starting at… 128.

243 00:20:16.330 00:20:21.879 Ashwini Sharma: Going back up to… I mean, like, any number of timeframes I can give, right?

244 00:20:27.440 00:20:31.459 Ashwini Sharma: Okay, so this is, going back up to 52 timeframes.

245 00:20:32.240 00:20:39.169 Ashwini Sharma: So it goes… since it is 1 week, it goes up to 52 weeks back, right? If I do 4 weeks, then it goes up to…

246 00:20:39.470 00:20:45.430 Ashwini Sharma: Pour into 10 bags, like, but all the records will be in that chunk.

247 00:20:45.720 00:20:46.410 Ashwini Sharma: Right?

248 00:20:48.480 00:20:53.309 Awaish Kumar: Give me data for 4 weeks, so it will be aggregated on 4 weeks already.

249 00:20:53.310 00:20:59.499 Ashwini Sharma: It’ll be aggregated in 4 weeks, and it will go back in time, right? That many time frames.

250 00:20:59.770 00:21:02.769 Ashwini Sharma: And then similarly, I can do it for 12 weeks, right?

251 00:21:03.870 00:21:04.700 Ashwini Sharma: Okay.

252 00:21:06.350 00:21:08.740 Ashwini Sharma: And then so on, like, 24, 52.

253 00:21:09.160 00:21:12.890 Ashwini Sharma: So, I am following this… this one. This is what I do.

254 00:21:14.790 00:21:27.029 Ashwini Sharma: Alright? Every time I send a request to this one, I tell it, give me data for weekly data going back up to 52 weeks. Because if I have weekly data, I can always

255 00:21:27.140 00:21:30.310 Ashwini Sharma: generate this one. I can always generate this one.

256 00:21:33.150 00:21:33.850 Awaish Kumar: Huh.

257 00:21:33.980 00:21:34.780 Ashwini Sharma: Okay?

258 00:21:35.210 00:21:37.370 Ashwini Sharma: Now, the issue here is…

259 00:21:37.670 00:21:43.470 Ashwini Sharma: When I sum it up, I get very close to getting data that I will get from this one.

260 00:21:49.610 00:21:50.830 Ashwini Sharma: And I’ve done that.

261 00:21:50.980 00:22:00.639 Ashwini Sharma: I extracted data using both these techniques, okay? And then I compared. Now, the Spins API is returning more data when I send requests like this.

262 00:22:01.230 00:22:02.969 Awaish Kumar: Yeah, I have a question now.

263 00:22:02.970 00:22:03.570 Ashwini Sharma: Yeah.

264 00:22:03.730 00:22:12.770 Awaish Kumar: In our first request, when you say one week grain, and going back 52 means we are going back, like, 52 weeks.

265 00:22:13.380 00:22:14.100 Ashwini Sharma: Yes.

266 00:22:14.890 00:22:17.529 Awaish Kumar: We are going to get 52 weeks of data, right?

267 00:22:17.530 00:22:18.190 Ashwini Sharma: Yes.

268 00:22:20.370 00:22:24.069 Awaish Kumar: And when I say guile is 4 weeks, that means I’m…

269 00:22:24.200 00:22:29.029 Awaish Kumar: I’m in a one request, I’m getting data, for a month.

270 00:22:29.520 00:22:35.539 Awaish Kumar: I’m actually getting… going back 10 months. That… that is covered in the…

271 00:22:36.220 00:22:40.229 Awaish Kumar: In the weekly one, because that’s already 52 weeks of data.

272 00:22:41.350 00:22:45.150 Awaish Kumar: Right? Def… That’s why we are able to cover it.

273 00:22:48.720 00:22:50.769 Ashwini Sharma: Sorry, I didn’t follow that. Can you repeat it again?

274 00:22:50.770 00:22:54.889 Awaish Kumar: So… Yeah, when you… you are saying that we have one week

275 00:22:55.970 00:23:04.739 Awaish Kumar: of data. One week gets a granularity, then if we get 52… if we get 52 rows, that means we have 52 weeks of data.

276 00:23:04.740 00:23:05.420 Ashwini Sharma: Yes.

277 00:23:06.160 00:23:09.530 Awaish Kumar: And similarly, if you have 1 month of data, 4 weeks is equal to…

278 00:23:09.880 00:23:10.400 Ashwini Sharma: Yeah.

279 00:23:10.600 00:23:11.540 Awaish Kumar: One month.

280 00:23:11.750 00:23:12.540 Ashwini Sharma: More or less, yeah.

281 00:23:12.540 00:23:18.949 Awaish Kumar: 10 is 1 month, then if there are 10 timeframes, then we are going back, like, around 10 months.

282 00:23:18.950 00:23:19.900 Ashwini Sharma: Yes.

283 00:23:20.530 00:23:24.439 Awaish Kumar: So, so if you get one week’s data.

284 00:23:24.810 00:23:30.490 Awaish Kumar: At 52 weeks, then basically that covers the The second request is that.

285 00:23:30.720 00:23:42.099 Ashwini Sharma: Yes, it will cover, but the issue is, like, this is what we’ll get, right? So, for example, PK1, I’m just saying this is primary K1, okay? Now, week is, 12-28.

286 00:23:42.630 00:23:46.469 Ashwini Sharma: Alright? And there is value 1, okay? This is what I’m getting.

287 00:23:46.710 00:23:50.580 Ashwini Sharma: Now, PK2, Let’s say 1228.

288 00:23:50.810 00:23:51.980 Ashwini Sharma: Value 2.

289 00:23:52.110 00:23:52.900 Ashwini Sharma: Okay.

290 00:23:53.280 00:23:57.329 Ashwini Sharma: Now, PK1, again, 12, 21.

291 00:23:57.580 00:23:58.819 Ashwini Sharma: Value 3.

292 00:23:59.020 00:24:00.690 Ashwini Sharma: PK2.

293 00:24:00.880 00:24:09.650 Ashwini Sharma: 121, value 3, okay? Now let’s take this example, right? This is what I’m getting when I’m sending this first request.

294 00:24:10.630 00:24:11.320 Awaish Kumar: Okay.

295 00:24:13.850 00:24:15.539 Ashwini Sharma: First request, right?

296 00:24:16.950 00:24:22.159 Ashwini Sharma: Now, when I’m sending this one, 4 weeks, second request.

297 00:24:26.330 00:24:28.920 Ashwini Sharma: I get something different.

298 00:24:30.620 00:24:34.999 Ashwini Sharma: Okay, wait, wait a second, this has to change a little bit, because,

299 00:24:35.900 00:24:40.800 Ashwini Sharma: So this has to go back up to 4 weeks, right? 121 minus 7 is 14.

300 00:24:42.280 00:24:45.390 Ashwini Sharma: 14, and then minus 7 is 7, right?

301 00:24:46.040 00:24:46.710 Ashwini Sharma: Okay.

302 00:24:48.150 00:24:48.940 Ashwini Sharma: Alright.

303 00:24:50.490 00:24:57.140 Ashwini Sharma: Now, when I do this thing, 28, value 1, value 2.

304 00:24:57.640 00:24:59.510 Ashwini Sharma: I get something like this.

305 00:25:06.710 00:25:08.390 Ashwini Sharma: You see the problem here?

306 00:25:08.920 00:25:16.819 Ashwini Sharma: Now, when I compare this value, right, this value plus this value plus this value plus this value.

307 00:25:16.970 00:25:18.280 Ashwini Sharma: with this value.

308 00:25:19.030 00:25:20.009 Awaish Kumar: Oh, it was not equal.

309 00:25:20.010 00:25:25.990 Ashwini Sharma: with basically some of these values, right? It’s… I’m comparing, like,

310 00:25:27.320 00:25:32.130 Ashwini Sharma: Yeah, for PK1, if I try to compare, right, 4 values and this one.

311 00:25:32.250 00:25:40.819 Ashwini Sharma: No, not exactly. So, this PK is at a… at a fine-grained level, right? This is not where they aggregate, right? They aggregate at…

312 00:25:40.950 00:25:47.089 Ashwini Sharma: Let’s say PKA plus PK1, right? This is basically primary key, okay?

313 00:25:48.490 00:25:54.200 Ashwini Sharma: So, so they aggregate at a higher level, and when they do that aggregation, that sum does not match.

314 00:25:55.050 00:25:55.800 Ashwini Sharma: Okay.

315 00:25:56.170 00:25:59.610 Ashwini Sharma: So that’s… that’s the… main problem.

316 00:25:59.930 00:26:06.409 Awaish Kumar: That’s… you are saying when, for the PK1, when I sum the 4 weeks of values.

317 00:26:06.470 00:26:09.159 Ashwini Sharma: It’s actually not equal to the…

318 00:26:09.700 00:26:12.570 Awaish Kumar: The volume which you are getting from 4-week request.

319 00:26:12.570 00:26:19.770 Ashwini Sharma: So, no, not at PK1 level, it will be at a higher level, right? So, this PK1 is a combination of

320 00:26:20.970 00:26:25.070 Ashwini Sharma: PG1 equals to, let’s say, Geography.

321 00:26:25.580 00:26:26.370 Awaish Kumar: Okay.

322 00:26:26.630 00:26:36.460 Ashwini Sharma: Geography plus, brand, brand plus category, product universe, right, plus reporting level.

323 00:26:37.010 00:26:37.740 Awaish Kumar: Okay.

324 00:26:38.120 00:26:39.950 Ashwini Sharma: Category plus subcategory.

325 00:26:41.900 00:26:46.539 Ashwini Sharma: Alright. Now, sometimes they do aggregation at geography level, right?

326 00:26:46.700 00:26:49.209 Ashwini Sharma: Geography level, let’s say. Yeah?

327 00:26:49.210 00:26:49.560 Awaish Kumar: Perfect.

328 00:26:49.560 00:26:51.710 Ashwini Sharma: Sometimes they do an aggregation at category.

329 00:26:52.320 00:26:56.590 Ashwini Sharma: Okay, and sometimes they do an aggregation at geographic. So, basically.

330 00:26:57.070 00:26:59.870 Ashwini Sharma: Geography is also there, part of. Alright.

331 00:27:00.610 00:27:05.610 Ashwini Sharma: Now, when you do this, multiple PKs will come within it, right?

332 00:27:07.960 00:27:19.450 Ashwini Sharma: See, when I do an aggregation, see, if I just take this entire thing, then I’ll get one value, right? But if I do an aggregation at geography level, then all other PKs which have

333 00:27:20.220 00:27:22.680 Ashwini Sharma: Different, this thing.

334 00:27:22.830 00:27:26.029 Ashwini Sharma: But same geography level, will fall under the same bucket.

335 00:27:30.940 00:27:31.760 Awaish Kumar: Okay.

336 00:27:31.760 00:27:41.600 Ashwini Sharma: Okay, so, yeah, and there is slight difference in the numbers that we get when I do an aggregation on geography level and this one.

337 00:27:47.150 00:27:51.230 Awaish Kumar: Okay, so, like, from… for the 4 weeks request.

338 00:27:51.760 00:27:57.000 Awaish Kumar: We… we are getting aggregated data on a geographical… geography level.

339 00:27:58.240 00:28:02.349 Awaish Kumar: So, what is PK1 in the other request, also? Is it the same, or…

340 00:28:02.350 00:28:09.690 Ashwini Sharma: Let me show you over here, because this is complex to explain, right? Okay, see here, right?

341 00:28:10.380 00:28:15.289 Ashwini Sharma: So this is for the same, right? These are two different data sets, okay?

342 00:28:15.290 00:28:18.580 Awaish Kumar: Like, do you… can you go back one second?

343 00:28:20.680 00:28:24.439 Awaish Kumar: Like, let’s call these requests 123. Let’s name them.

344 00:28:24.920 00:28:25.800 Ashwini Sharma: Okay.

345 00:28:28.570 00:28:30.920 Awaish Kumar: One of these records, yeah, it is. It’s called…

346 00:28:31.670 00:28:33.899 Awaish Kumar: Just the beginning, you can add one dot.

347 00:28:34.080 00:28:35.460 Awaish Kumar: to not clear. Okay.

348 00:28:40.210 00:28:48.450 Awaish Kumar: So, I’m saying, in the first request, you mentioned, yeah, some PK1, which are basically these columns.

349 00:28:48.660 00:28:53.520 Awaish Kumar: Which… if we… which gives us the… For… for this…

350 00:28:54.570 00:28:57.750 Awaish Kumar: Primarily, we have a week’s of data.

351 00:28:58.020 00:28:59.249 Awaish Kumar: One week of data.

352 00:28:59.440 00:29:06.469 Awaish Kumar: So, from second request, we have 4 vCompleteer, but is the primary key same for the second request?

353 00:29:06.800 00:29:17.499 Ashwini Sharma: Yeah, primary key is the same, right? But there… yeah, that’s what… there is a possibility that they are going to add more primary keys, right? That’s what I’m trying to show you over here, right?

354 00:29:18.230 00:29:20.089 Ashwini Sharma: Take a look at this one.

355 00:29:21.690 00:29:26.580 Ashwini Sharma: Let’s see here, see, let’s take a look at this one, RMA, right?

356 00:29:27.060 00:29:31.370 Awaish Kumar: I’m just asking you, do you want to join CTA, or… what do I’m saying?

357 00:29:31.370 00:29:33.620 Ashwini Sharma: Yeah, yeah, I have to join CTA also.

358 00:29:33.620 00:29:35.469 Awaish Kumar: We have a call now, let’s get back to it.

359 00:29:35.470 00:29:37.940 Ashwini Sharma: Yeah, let’s, yeah.

360 00:29:38.520 00:29:46.259 Ashwini Sharma: Okay, I’ll drop mid-CTA, and then make this document, right? Because we have to send it out today.

361 00:29:46.610 00:29:50.369 Ashwini Sharma: And then we can sync back later, once the CTA meeting is over.

362 00:29:51.070 00:29:53.670 Awaish Kumar: Okay, when is the Magic Spoon meeting?

363 00:29:54.100 00:29:57.479 Ashwini Sharma: Magic Spoon meeting is… I think, I think it’s right out of CTA.

364 00:29:58.570 00:29:59.220 Awaish Kumar: Oh, cool.

365 00:29:59.910 00:30:00.610 Ashwini Sharma: Yeah.

366 00:30:01.780 00:30:08.129 Ashwini Sharma: Yeah, half an hour after CTA it is, so I’m going to drop somewhere middle of the CTA only.

367 00:30:09.320 00:30:10.120 Ashwini Sharma: Alright.

368 00:30:10.230 00:30:11.760 Ashwini Sharma: Well, let’s, let’s join there.