Meeting Title: AI Service Daily Recap - Blockers + Realign Date: 2026-04-30 Meeting participants: Samuel Roberts, Mustafa Raja, Casie Aviles, Pranav Narahari


WEBVTT

1 00:00:10.880 00:00:11.690 Mustafa Raja: Hey.

2 00:00:14.700 00:00:15.380 Samuel Roberts: Boo.

3 00:00:16.110 00:00:16.899 Mustafa Raja: How are you?

4 00:00:17.950 00:00:23.190 Samuel Roberts: Doing okay, eating some… Roadblocks with my open code, but, you know.

5 00:00:23.370 00:00:25.150 Samuel Roberts: It’s what’s new.

6 00:00:29.510 00:00:30.740 Samuel Roberts: Eden was gone.

7 00:00:31.500 00:00:32.600 Samuel Roberts: Yeah, good. Sorry?

8 00:00:33.330 00:00:41.350 Mustafa Raja: GLM was nice. I started using VLM as, My, you know, go-to…

9 00:00:42.620 00:00:44.529 Mustafa Raja: model for open code, and it’s been.

10 00:00:44.530 00:00:44.910 Samuel Roberts: Oh, yeah.

11 00:00:45.480 00:00:46.230 Mustafa Raja: Right now.

12 00:00:46.630 00:00:50.060 Samuel Roberts: Oh, nice. Yeah, I haven’t had a chance to try it yet.

13 00:00:52.000 00:00:54.000 Mustafa Raja: Yeah, it’s nice. I recommend it.

14 00:00:54.350 00:00:55.299 Samuel Roberts: And I’ll try it out.

15 00:00:59.090 00:01:00.840 Samuel Roberts: How about you, Casey, how are things?

16 00:01:02.430 00:01:05.160 Casie Aviles: Oh, hey, yeah, I was just,

17 00:01:05.960 00:01:14.860 Casie Aviles: working on Eden, and I asked, like, an initial, broad Question to the… Agent.

18 00:01:15.670 00:01:16.340 Samuel Roberts: Nice.

19 00:01:19.930 00:01:20.500 Samuel Roberts: Cool.

20 00:01:22.420 00:01:28.870 Casie Aviles: I’ll wait a little for Prano to kind of go in depth, but I guess what I was thinking is, like.

21 00:01:30.050 00:01:33.880 Casie Aviles: I guess we need to understand, like, what…

22 00:01:34.440 00:01:39.260 Casie Aviles: what exactly the… so the COO will be using this, right? Primarily, the COO.

23 00:01:39.260 00:01:40.580 Samuel Roberts: Correct, yeah.

24 00:01:41.990 00:01:45.019 Casie Aviles: Yeah, so kind of understanding, like, what… what…

25 00:01:45.620 00:01:48.799 Casie Aviles: The report is supposed to do for them, you know, like…

26 00:01:49.430 00:01:55.230 Casie Aviles: because right now, like, the summaries that I… or the themes that I generated are more… more of just

27 00:01:55.340 00:01:59.059 Casie Aviles: Topic summaries, like, what’s happening across the org, so…

28 00:01:59.560 00:02:00.360 Samuel Roberts: Right.

29 00:02:01.800 00:02:04.850 Casie Aviles: If they have, like, other needs or other questions, like.

30 00:02:05.330 00:02:15.239 Casie Aviles: what needs my attention, or something that’s probably more actionable, then I guess we need to kind of think of… think of a taxonomy, I guess.

31 00:02:16.870 00:02:19.760 Casie Aviles: But yeah, that’s just some of my ideas.

32 00:02:21.840 00:02:22.750 Samuel Roberts: Hippen off.

33 00:02:24.950 00:02:30.050 Pranav Narahari: Hey guys, Casey, we were just talking about, kind of, the theme… Generation.

34 00:02:30.050 00:02:31.089 Casie Aviles: Oh, yeah.

35 00:02:31.090 00:02:37.070 Pranav Narahari: Okay, gotcha. And so, yeah, you were saying right now, mostly just…

36 00:02:37.800 00:02:40.649 Pranav Narahari: It’s basically turning out to just be, like, summaries.

37 00:02:44.140 00:02:47.509 Casie Aviles: So, yeah, it’s more of, like, a topic summary.

38 00:02:47.630 00:02:50.580 Casie Aviles: Yeah, let me quickly share.

39 00:02:50.900 00:02:53.490 Casie Aviles: As well, so it’s more visual.

40 00:02:58.970 00:03:05.510 Casie Aviles: Okay, so I have this ticket. Let me know if you guys aren’t seeing it, but I have this, thread with…

41 00:03:06.630 00:03:08.449 Casie Aviles: Oh, oops, I got logged out, but…

42 00:03:08.900 00:03:13.179 Casie Aviles: Essentially, I just found this first pass teams, so I have 5…

43 00:03:13.650 00:03:19.570 Casie Aviles: But these are primarily just about, like, what’s happening across the org.

44 00:03:21.410 00:03:22.040 Pranav Narahari: Yeah.

45 00:03:22.040 00:03:25.190 Casie Aviles: So… that’s what I have right now, but…

46 00:03:25.750 00:03:28.089 Casie Aviles: I think what, what…

47 00:03:28.400 00:03:31.939 Casie Aviles: We can do as well is, like…

48 00:03:32.560 00:03:42.780 Casie Aviles: understand, like, what the report should do for the COO. So, if they just want to know, like, what’s happening, then I guess something like this should be fine, although this would be, like.

49 00:03:43.240 00:03:48.089 Casie Aviles: different, likely different every week, right? So, there may be different…

50 00:03:48.760 00:03:52.250 Casie Aviles: Summaries like this, or topic summaries.

51 00:03:52.440 00:03:54.559 Casie Aviles: But if there are questions… yeah.

52 00:03:55.200 00:03:57.999 Pranav Narahari: Sorry, yeah, I was just gonna say that…

53 00:03:58.650 00:04:18.239 Pranav Narahari: This is actually great. I don’t think… I think this would actually be super useful as well, just to see, like, week over week, how are things progressing? And so the idea overall was… would be like, hey, yeah, we have this report, let’s save it, let’s see how it compares to the report that’s generated the next Friday, for example. However, I think what would…

54 00:04:18.329 00:04:30.270 Pranav Narahari: what Danny has said that is, something that’s super useful for him is to just get an understanding of what are the bottlenecks currently across all projects.

55 00:04:30.840 00:04:32.260 Pranav Narahari: And so…

56 00:04:32.960 00:04:43.300 Pranav Narahari: I don’t know, I don’t… I thought we defined that somewhere in, like, a linear ticket, or maybe in, like, a Notion doc. But yeah, if you haven’t seen any reference of, like.

57 00:04:43.630 00:04:48.069 Pranav Narahari: bottlenecks or anything, then I can try to dig up that or write something.

58 00:04:48.070 00:04:55.390 Casie Aviles: Yeah, I do remember seeing that, but when I was checking the repo right now, I think…

59 00:04:56.240 00:05:01.590 Casie Aviles: the AI wasn’t able to pick it up, so I was wondering… If…

60 00:05:01.980 00:05:06.649 Casie Aviles: I just wrongly remembered it, or if that actually existed.

61 00:05:07.200 00:05:10.229 Pranav Narahari: Yeah, no, it should exist,

62 00:05:10.710 00:05:16.850 Pranav Narahari: What is that bandwidth and friction, like, for Milestone 6? Bandwidth and friction signals report?

63 00:05:18.900 00:05:20.060 Casie Aviles: Oh, okay.

64 00:05:20.170 00:05:21.620 Casie Aviles: bandwidth and so…

65 00:05:31.420 00:05:32.300 Pranav Narahari: Yeah.

66 00:05:32.410 00:05:37.970 Pranav Narahari: basically just kind of the three things that were just mentioned up there. So, like… .

67 00:05:37.970 00:05:38.930 Casie Aviles: Oh, okay.

68 00:05:39.160 00:05:39.980 Pranav Narahari: Yeah.

69 00:05:40.580 00:05:47.169 Pranav Narahari: So, like, talk to task is basically how, like, the amount of meetings, the amount of, like.

70 00:05:47.280 00:05:51.960 Pranav Narahari: communications that are happening, to just…

71 00:05:52.150 00:05:58.759 Pranav Narahari: exchange information before actually building anything, how much of that is happening? So…

72 00:05:58.860 00:06:05.930 Pranav Narahari: I mean, I think you probably built out, if you were able to generate the summaries, it’s probably just another level of prompting.

73 00:06:06.550 00:06:08.610 Pranav Narahari: to… get the…

74 00:06:08.610 00:06:09.160 Casie Aviles: What?

75 00:06:09.160 00:06:18.890 Pranav Narahari: specifics of those… for, like, specifics of bottlenecks, you know, friction indicators, things like that, right? Or do you think this is, like, a totally different ask?

76 00:06:20.980 00:06:29.720 Casie Aviles: I think we can definitely just start with prompting first, and then we can build on top of that if we need to, but yeah, that should be good.

77 00:06:30.150 00:06:31.370 Pranav Narahari: Yeah, okay, cool.

78 00:06:32.640 00:06:33.440 Pranav Narahari: Okay.

79 00:06:33.720 00:06:34.819 Pranav Narahari: Sounds good.

80 00:06:37.980 00:06:45.439 Casie Aviles: I guess what… my last question is more on the technical side. So the report will just be, like, a…

81 00:06:46.260 00:06:52.950 Casie Aviles: Like, you guys mentioned, the deterministic, so that would be sent out similar to, like, just the daily memos we’re sending.

82 00:06:56.960 00:06:57.280 Pranav Narahari: That would be…

83 00:06:57.280 00:07:01.360 Casie Aviles: for, like, ABC, for example, like, something like that, right?

84 00:07:02.930 00:07:05.849 Pranav Narahari: Yeah, I think deterministic in just that…

85 00:07:06.810 00:07:24.979 Pranav Narahari: we don’t want to specify, like, okay, always look into these projects, because, you know, projects are going to be ever-changing, and I don’t think it’ll capture everything going on in the organization that way. I was saying, just, like, deterministic, maybe, in, like, the format.

86 00:07:25.240 00:07:25.800 Pranav Narahari: Okay.

87 00:07:25.800 00:07:26.500 Casie Aviles: Yeah, okay.

88 00:07:26.500 00:07:37.060 Pranav Narahari: going to be something that is automated and generated week over week. There should probably be a section of assessing, okay, these are blockers that we noticed last week, have we noticed improvements or declined?

89 00:07:37.060 00:07:37.740 Samuel Roberts: Yeah.

90 00:07:37.740 00:07:41.870 Pranav Narahari: Just, like, simple things like that, hmm.

91 00:07:42.940 00:07:49.479 Pranav Narahari: But other than that, I think it’s basically what you did here, like, creating summaries,

92 00:07:49.640 00:07:53.969 Pranav Narahari: Just kind of creating sections for the individual…

93 00:07:54.150 00:07:57.009 Pranav Narahari: individual portions of the report, and those…

94 00:07:57.950 00:08:05.680 Pranav Narahari: I think there’d probably be, like, those 3… 3 things that was just mentioned in Milestone 6. Like, the bandwidth, friction… Yeah.

95 00:08:06.340 00:08:09.739 Pranav Narahari: you know, those are gonna be some three questions, and let the AI just kind of, like.

96 00:08:10.490 00:08:16.050 Pranav Narahari: With that information, find relevant data within the Eden org.

97 00:08:18.280 00:08:18.930 Casie Aviles: Okay.

98 00:08:19.470 00:08:20.000 Pranav Narahari: No.

99 00:08:20.560 00:08:21.679 Pranav Narahari: Does that kind of…

100 00:08:22.300 00:08:33.659 Pranav Narahari: Let me know if that’s, like, not clear, like, we can work that out a little bit more. I think that’s probably enough to kind of just get an initial POC, and then we can kind of work to refine that going forward.

101 00:08:35.080 00:08:37.480 Casie Aviles: Okay. Yeah, that makes sense.

102 00:08:40.439 00:08:43.439 Pranav Narahari: Okay, cool. How about on…

103 00:08:43.579 00:08:48.189 Pranav Narahari: For Mustafa on the other Eden stuff that you’re working on?

104 00:08:48.740 00:08:53.590 Mustafa Raja: So… I have done the,

105 00:08:54.540 00:09:12.569 Mustafa Raja: backfill for all of the participants that are in the roster, and then, there were some, failures, so I’m taking a look at that, and then, the next step is going to be setting up a scheduler in GCP, so we can run it on a schedule.

106 00:09:13.520 00:09:17.409 Mustafa Raja: Every day. So, let me know what’s a good SLA for this, you know?

107 00:09:17.640 00:09:19.479 Mustafa Raja: Is one today good enough?

108 00:09:19.840 00:09:21.290 Pranav Narahari: Yeah, once a day is good.

109 00:09:22.020 00:09:26.879 Mustafa Raja: Okay, so I’ll run it on the start of the day, if that’s fine.

110 00:09:27.240 00:09:30.040 Samuel Roberts: I would do it, like, middle of the night, maybe, for…

111 00:09:30.500 00:09:34.110 Samuel Roberts: That way we get, like, all the previous day’s stuff, and it’s all in there.

112 00:09:34.340 00:09:34.900 Pranav Narahari: That works.

113 00:09:35.290 00:09:37.110 Mustafa Raja: That makes sense. Yeah.

114 00:09:37.110 00:09:38.200 Samuel Roberts: There, yeah.

115 00:09:38.350 00:09:38.870 Pranav Narahari: Yeah.

116 00:09:40.440 00:09:44.290 Pranav Narahari: Like, midnight mountain time, or 2 a.m. Mountain Time?

117 00:09:44.290 00:09:45.050 Samuel Roberts: Yeah, something like that.

118 00:09:45.050 00:09:45.560 Mustafa Raja: Certain time.

119 00:09:47.980 00:09:51.770 Pranav Narahari: Mountain Time is 1 hour behind Central.

120 00:09:52.400 00:09:56.330 Mustafa Raja: Okay, yeah, okay, yeah, I can do that. Yeah, but plus 7 or something.

121 00:09:57.050 00:10:01.639 Mustafa Raja: Yeah, and then Cloud Run was facing some issues,

122 00:10:01.880 00:10:10.040 Mustafa Raja: I fixed that, and then, a co-pilot kit, wasn’t working as expected. It’s now fine.

123 00:10:10.420 00:10:11.010 Pranav Narahari: Cool.

124 00:10:12.250 00:10:14.119 Mustafa Raja: That’s pretty much it.

125 00:10:15.270 00:10:18.490 Mustafa Raja: So I’m close to setting up the schedule.

126 00:10:20.040 00:10:21.230 Pranav Narahari: Okay, cool.

127 00:10:22.750 00:10:29.240 Samuel Roberts: Cool, yeah, transcripts, after…

128 00:10:30.150 00:10:33.089 Mustafa Raja: Yeah, I said, let me know if there’s anything else after that.

129 00:10:34.800 00:10:39.430 Pranav Narahari: Yeah, I think that’s good for right now. Just, let us know when those are done.

130 00:10:39.430 00:10:40.000 Samuel Roberts: Yeah.

131 00:10:40.320 00:10:40.870 Pranav Narahari: Yeah.

132 00:10:41.760 00:10:42.950 Pranav Narahari: And then, yes, I’m sure.

133 00:10:43.760 00:10:55.410 Mustafa Raja: I’m logging my time, so is this… is this going to be the same as the other Eden in Clockify? Like, is it the same project, or something else?

134 00:10:55.410 00:10:56.959 Samuel Roberts: This is an Eden AI project.

135 00:10:56.960 00:10:57.450 Pranav Narahari: Yeah.

136 00:10:57.590 00:11:03.290 Mustafa Raja: Let me see if I am in that, then. Because I’ve been logging in regularly.

137 00:11:03.520 00:11:10.060 Samuel Roberts: Oh, yeah, definitely update that if you… if you don’t have it, definitely get, put a request into ops, and then update those.

138 00:11:11.240 00:11:14.669 Mustafa Raja: Yeah, I’ve been adding it in the evening hours.

139 00:11:16.940 00:11:18.570 Samuel Roberts: Especially by the end of the month.

140 00:11:19.260 00:11:23.620 Mustafa Raja: Oh, yeah, I’m only needing health. Okay, I’ll request health.

141 00:11:23.890 00:11:24.460 Samuel Roberts: Okay.

142 00:11:25.340 00:11:25.840 Mustafa Raja: Thank you.

143 00:11:25.840 00:11:27.039 Samuel Roberts: Good catch, good catch.

144 00:11:29.670 00:11:38.739 Samuel Roberts: Yeah, transcripts, I spent a little while trying to figure out if they were able to do tags for any of those cancellations, and I wasn’t able to find anything with the API.

145 00:11:39.050 00:11:44.780 Samuel Roberts: But, I was able to pull the window ones. There’s not a ton of those, so it was kind of easy to…

146 00:11:44.930 00:11:46.669 Samuel Roberts: To pull… er, quick to pull down.

147 00:11:46.820 00:11:48.170 Samuel Roberts: I’m…

148 00:11:48.440 00:11:56.050 Samuel Roberts: I did, like, a quick pass of finding some of the cancellation ones, but my next step is to actually put together, like, a… an agent that will actually just systematically

149 00:11:56.280 00:12:12.710 Samuel Roberts: analyze the transcript. Like, I gotta fetch the… you know, metadata, fetch the transcript, compile the transcript, pass that through an agent, analyze if that’s a cancellation. So I did kind of the basic work to make sure that that is feasible, and now I’m gonna start putting together, like.

150 00:12:12.730 00:12:24.569 Samuel Roberts: a master agent, and probably, like, a little UI to see it, which ones are cancellations in the last 14 days, stuff like that. So, hopefully… I mean, not hopefully, it will be able to be adjusted for the…

151 00:12:24.570 00:12:33.089 Samuel Roberts: different queues, I think, is probably the best way to do it. And then when we need to add more, we can just change that and have it, like, rerun.

152 00:12:33.350 00:12:33.880 Samuel Roberts: So…

153 00:12:34.410 00:12:37.419 Pranav Narahari: Yeah, that’s why I’m not too concerned that we don’t know exactly which…

154 00:12:37.700 00:12:38.710 Samuel Roberts: Yeah.

155 00:12:38.840 00:12:42.710 Pranav Narahari: Because it’s as simple as probably just updating an environment variable.

156 00:12:42.710 00:13:00.500 Samuel Roberts: Yeah, yeah, so I mean, the other side of this is, like, we could just be ingesting them all, and putting them into BQ, and then analyzing all that later, but that seems like a lot of work for what we’re getting out of it right now. So I feel like that’s, you know, I don’t want to over-optimize kind of thing. I think this is a good plan for now.

157 00:13:00.760 00:13:02.550 Pranav Narahari: Exactly. Yep. Cool. Cool.

158 00:13:02.670 00:13:03.450 Samuel Roberts: Okay.

159 00:13:04.200 00:13:11.340 Pranav Narahari: Awesome. Sounds like we’re in a pretty good spot. Yeah, so Casey, Sam, you guys are gonna be out tomorrow.

160 00:13:12.880 00:13:21.499 Pranav Narahari: Tomorrow for Eden is what I’m probably most concerned about. Even that, I don’t think the call’s gonna end up happening.

161 00:13:21.650 00:13:27.639 Pranav Narahari: Yeah, actually, Caitlin just messaged me, it is not going to be happening. Okay,

162 00:13:27.970 00:13:33.780 Pranav Narahari: So… I’m not too concerned. Yeah, I think we’re in a pretty good spot.

163 00:13:34.780 00:13:41.820 Pranav Narahari: if anything, you know, I’ll be around, too, so I can, help. We can figure something out together, Mustaf and I.

164 00:13:42.370 00:13:54.509 Samuel Roberts: Yeah, I would say, take a look at the dev deployment, now that Mustafa’s got that working, and just, you know, that’s something you can demo or record, or, you know, it’s got the compare, the Eden dash compare.

165 00:13:54.680 00:13:58.629 Samuel Roberts: And the Eden is the co-pilot one, so there’s two URLs there.

166 00:13:58.870 00:14:05.849 Mustafa Raja: Yeah, I’m also wondering if you want to, you know, push it to QA or even main?

167 00:14:06.250 00:14:09.580 Mustafa Raja: You know, when is a good time to do that, you know?

168 00:14:13.580 00:14:15.439 Samuel Roberts: Maybe we can do that, exactly.

169 00:14:15.610 00:14:16.220 Pranav Narahari: I would say…

170 00:14:16.220 00:14:21.999 Samuel Roberts: I wouldn’t worry about it too much until there’s some more testing in dev, maybe. Yeah. But I think it’s…

171 00:14:22.920 00:14:36.289 Pranav Narahari: Yeah, so I totally agree. And Sam, for ABC, if you could actually just, do a quick, like, Zoom clip again for, just, like, the transcript stuff, just kind of showing,

172 00:14:37.520 00:14:41.580 Pranav Narahari: Maybe, actually, what we can do is,

173 00:14:41.880 00:15:00.150 Pranav Narahari: is there any type of showcase that we can do there that I… like, for showing this tomorrow to the client? I just want to show them, like, how we’ve kind of thought about things in terms of ingesting the transcripts, and then finally, like, making them, like, differentiate between, like, canceled or not canceled content.

174 00:15:00.150 00:15:00.770 Samuel Roberts: Yeah.

175 00:15:00.770 00:15:01.260 Pranav Narahari: And then…

176 00:15:01.260 00:15:01.650 Samuel Roberts: I’m…

177 00:15:01.650 00:15:07.010 Pranav Narahari: I don’t know if you’ve gotten also to the point of, like, putting them into individual buckets for why they’ve been canceled.

178 00:15:07.540 00:15:22.490 Samuel Roberts: No, no, not yet. There were literally, like, 10 that it identified as cancellations for those ones, so I… it wasn’t, like, I didn’t dive too deep into that yet. I figured that would be after I… because some of them were, like, false positives and false negatives, so I’m trying to… I want to get it a little better than just dumping it into…

179 00:15:22.720 00:15:23.470 Samuel Roberts: Open code.

180 00:15:23.910 00:15:33.410 Pranav Narahari: So, I guess, actually, the main thing, we don’t really even need a video, because I just want to be able to talk about it tomorrow. Seems like whatever video you make, it’d be kind of too technical for them anyways.

181 00:15:33.410 00:15:38.249 Samuel Roberts: That’s one of the reasons I was hoping to put together, like, some little UI that you could just kind of see and visualize it, which.

182 00:15:38.250 00:15:44.649 Pranav Narahari: It wouldn’t be that useful for them right now, to be honest. Like, I don’t want to waste your time on that.

183 00:15:44.650 00:15:51.379 Samuel Roberts: Well, I think it’s actually going to be kind of helpful for me managing some of the stuff, because I’m getting lots of scripts and everything, and I just need something a little more, you know…

184 00:15:51.380 00:15:53.400 Pranav Narahari: That’s a good thing, yeah, that makes sense.

185 00:15:53.400 00:15:59.750 Samuel Roberts: But that’s what I was gonna say, is I might not get to, like, a good recording point yet, but I’ll make sure to have, like, a report, at least.

186 00:16:00.280 00:16:15.969 Pranav Narahari: Yeah, I was gonna say, I don’t even think you need to generate anything. I think I just want to know where we’re currently at, which I think I understand, is, you know, we’re able to ingest them, we haven’t built it to be automated, you know, on a weekly or daily cadence. However,

187 00:16:15.970 00:16:20.350 Pranav Narahari: We’ve ingested based on this specific tag, which is, Windows, currently.

188 00:16:20.350 00:16:22.630 Samuel Roberts: Yeah, the queue, it’s a queue, specifically.

189 00:16:22.630 00:16:24.560 Pranav Narahari: Oh, it’s not a tag, it’s a queue, okay.

190 00:16:24.560 00:16:35.129 Samuel Roberts: Yeah, I was having issues with tags, I’m not even sure what I’m getting right now that are tags, but the queue list that I sent you was something that Yvette had shared a while back that did match up to 8x8.

191 00:16:35.130 00:16:38.449 Pranav Narahari: Yeah, in Nicole, basically what I’m gonna do is, like.

192 00:16:38.670 00:16:45.359 Pranav Narahari: Yvette is gonna pull up 8x8, she’s gonna show me the queue list, and then she’s gonna basically point me to either a different

193 00:16:45.580 00:16:47.999 Pranav Narahari: Tag, or a different,

194 00:16:48.520 00:16:50.750 Pranav Narahari: like, enum in there, I guess?

195 00:16:50.760 00:17:02.469 Pranav Narahari: Or she’s going to point me to, hey, we don’t follow queues, actually, we want to follow, like, tags, and then, in which case, we’ll do something next week to sort out how can we pull a transcript based on that.

196 00:17:02.470 00:17:15.939 Pranav Narahari: And then, you know, I’ll also get information from her based on, like, okay, so this is what we were able to find, like, and we already built out the process for differentiating canceled versus non-canceled transcripts. What are the individual buckets within canceled now?

197 00:17:16.099 00:17:25.269 Pranav Narahari: Right. That you want us to, like, further filter each of these transcripts into. And, yeah, so I think that’s all I need. I just wanted to make sure, like,

198 00:17:25.650 00:17:27.050 Pranav Narahari: That sounds right to you.

199 00:17:27.680 00:17:37.590 Samuel Roberts: Yeah, yeah, it does. I think, yeah, the queues that you talked about before, I think it’s, like, where it gets routed, basically. So that should be, like, relatively department-specific, I think.

200 00:17:38.580 00:17:48.259 Samuel Roberts: I just… it didn’t quite map to, like, home improvement or home whatever for washing and stuff, so I just… a little clarification there, and then whichever ones they want us to dig into would be good.

201 00:17:48.260 00:17:49.060 Pranav Narahari: From that.

202 00:17:49.220 00:17:50.669 Samuel Roberts: I think, yeah, I think it’s perfect.

203 00:17:52.750 00:17:53.110 Pranav Narahari: Cool.

204 00:17:53.110 00:17:53.650 Samuel Roberts: Boop.

205 00:17:54.160 00:17:56.520 Pranav Narahari: Awesome, guys, I think we’re… I think we’re good to go, then.

206 00:17:56.800 00:17:57.650 Samuel Roberts: Alrighty.

207 00:17:58.200 00:17:59.060 Samuel Roberts: Thanks, Bill.

208 00:17:59.740 00:18:01.820 Pranav Narahari: Thanks, guys. Talk soon.

209 00:18:01.820 00:18:02.520 Casie Aviles: Thank you.