Meeting Title: ABC Spreadsheet Data Validation Check Date: 2025-07-01 Meeting participants: Amber Lin, Casie Aviles, Uttam Kumaran
WEBVTT
1 00:00:05.210 ⇒ 00:00:07.190 Amber Lin: Just for us, but just for us.
2 00:11:13.030 ⇒ 00:11:13.720 Amber Lin: Hi!
3 00:11:14.800 ⇒ 00:11:17.459 Casie Aviles: Yeah, we were just in our stand up.
4 00:11:17.980 ⇒ 00:11:18.680 Amber Lin: Okay.
5 00:11:20.210 ⇒ 00:11:22.689 Casie Aviles: Yeah. Autumn should be joining, I think, yeah.
6 00:11:22.976 ⇒ 00:11:26.129 Amber Lin: Okay, sounds good. I don’t think this will take too long.
7 00:11:27.180 ⇒ 00:11:27.790 Casie Aviles: Okay.
8 00:11:45.430 ⇒ 00:11:47.049 Casie Aviles: Should I share my screen or.
9 00:11:47.664 ⇒ 00:11:51.895 Amber Lin: Sure I can. I can share my screen now. I’ll just walk you through real quick.
10 00:11:52.620 ⇒ 00:11:52.980 Casie Aviles: Okay.
11 00:11:52.980 ⇒ 00:12:03.150 Amber Lin: If so, we’ve already flattened the spreadsheet, and we just want to check with you
12 00:12:03.400 ⇒ 00:12:08.120 Amber Lin: what you think about it, and then how we can do the data validation.
13 00:12:10.120 ⇒ 00:12:13.130 Amber Lin: So 1st off this spreadsheet is set up.
14 00:12:14.460 ⇒ 00:12:16.749 Amber Lin: Each row is a different zip code.
15 00:12:17.070 ⇒ 00:12:31.020 Amber Lin: and then each column is a different service, and this service is either under residential or it’s under pets, and it’s residential. It requires a specialty technician, but we have it in terms of granularity
16 00:12:31.886 ⇒ 00:12:35.760 Amber Lin: of each, each individual service.
17 00:12:36.520 ⇒ 00:12:39.069 Amber Lin: So that’s how this main spreadsheet is set up.
18 00:12:39.390 ⇒ 00:12:47.529 Amber Lin: And then how we did. This is, 1st of all, let’s look at the original inspector sheets.
19 00:12:48.380 ⇒ 00:13:04.699 Amber Lin: So generally they look like this. So they have groupings of Zip code. And then they talk about okay, this specific inspector services, these zip codes, and they can do commercial or residential, and they have other small
20 00:13:04.810 ⇒ 00:13:09.379 Amber Lin: requirements that may or may not apply to all of them.
21 00:13:09.870 ⇒ 00:13:11.479 Amber Lin: And this is
22 00:13:11.610 ⇒ 00:13:29.430 Amber Lin: the case for the different zip codes, and they also have it for services such as, Okay, Chem, 3 has its specific definitions, and Handyman has its definitions. So each of them is separated by
23 00:13:31.270 ⇒ 00:13:36.460 Amber Lin: say, by person. So this person services the zip codes for this area.
24 00:13:36.840 ⇒ 00:13:40.137 Amber Lin: and what we did is say, let’s take
25 00:13:41.860 ⇒ 00:13:45.170 Amber Lin: Let’s take Austin residential as an example.
26 00:13:45.310 ⇒ 00:13:53.630 Amber Lin: Alright. So we broke this down by zip codes, and then, for, say.
27 00:13:54.435 ⇒ 00:14:02.180 Amber Lin: residential pest, we added, all these inspectors that does that for wdi.
28 00:14:02.820 ⇒ 00:14:08.549 Amber Lin: We added these 3 people that says they can do wdi. We added it there
29 00:14:08.930 ⇒ 00:14:21.020 Amber Lin: and then for mosquitoes same thing based on what they said. All of them can do mosquitoes. So we say, yes, okay. All of them can do mosquitoes. Similarly, for the other services.
30 00:14:21.270 ⇒ 00:14:23.839 Amber Lin: such as bedbug that was mentioned.
31 00:14:24.696 ⇒ 00:14:25.830 Amber Lin: Up here.
32 00:14:27.200 ⇒ 00:14:30.910 Amber Lin: And after we did this spreadsheet for each of these tabs
33 00:14:31.110 ⇒ 00:14:40.110 Amber Lin: we went back. We made sure that the titles match this here.
34 00:14:40.490 ⇒ 00:14:48.910 Amber Lin: And then we did a essentially we did a vlookup with a match on titles.
35 00:14:49.730 ⇒ 00:14:52.179 Amber Lin: and that’s how we arrived at this sheet.
36 00:14:55.840 ⇒ 00:14:56.730 Uttam Kumaran: Okay.
37 00:15:03.470 ⇒ 00:15:06.850 Amber Lin: So what do you think? Is this a
38 00:15:07.871 ⇒ 00:15:12.500 Amber Lin: very error. Prone method, relatively. Okay. Method.
39 00:15:12.970 ⇒ 00:15:16.820 Amber Lin: How will we approach data validation for this.
40 00:15:23.710 ⇒ 00:15:26.570 Uttam Kumaran: Yeah. So I mean the way you would standard like.
41 00:15:27.240 ⇒ 00:15:33.729 Uttam Kumaran: Typically the way we would handle something like this. I’m just looking at the thing on my screen, the way you would handle something like this is
42 00:15:35.510 ⇒ 00:15:38.809 Uttam Kumaran: we would basically take the
43 00:15:41.390 ⇒ 00:15:48.010 Uttam Kumaran: the input request and run a query on a database that brings in the right value.
44 00:15:48.700 ⇒ 00:15:56.089 Uttam Kumaran: You guys tested this with a few examples. And I assume you’re just basically Casey bringing this into context when needed.
45 00:15:56.680 ⇒ 00:15:57.370 Uttam Kumaran: Or how is.
46 00:15:57.370 ⇒ 00:15:58.080 Casie Aviles: I know.
47 00:15:58.080 ⇒ 00:16:00.530 Uttam Kumaran: How has this sort of been working so far?
48 00:16:01.510 ⇒ 00:16:05.730 Casie Aviles: No, this hasn’t been integrated with the AI yet. So we were just focusing on.
49 00:16:05.730 ⇒ 00:16:06.340 Uttam Kumaran: Flat.
50 00:16:06.340 ⇒ 00:16:09.589 Casie Aviles: It and standardizing the data.
51 00:16:13.610 ⇒ 00:16:16.740 Casie Aviles: But yeah, that’s ideally how we would do it. We would also.
52 00:16:16.740 ⇒ 00:16:24.029 Uttam Kumaran: It’s only 700 rows. So I’m not like too opposed to actually throwing this into the Llm.
53 00:16:25.457 ⇒ 00:16:28.170 Uttam Kumaran: I guess my only question would be
54 00:16:32.090 ⇒ 00:16:39.569 Uttam Kumaran: like. How are they going to maintain this going forward.
55 00:16:40.610 ⇒ 00:16:46.529 Amber Lin: Yeah, that was also what I was concerned about. I think for that.
56 00:16:46.780 ⇒ 00:16:57.174 Amber Lin: first, st i need to get confirmation that they’re gonna use this sheet instead of their spreadsheet, which I think they will. If I was very excited to hear about this.
57 00:16:57.510 ⇒ 00:16:59.860 Uttam Kumaran: To give you a sense of what is wrong about theirs.
58 00:17:00.640 ⇒ 00:17:09.679 Uttam Kumaran: And like kind of the way to articulate, and I’ll and I’ll share, and I can just kind of give you. You’ll have this recording. But I’ll kind of give you like what is
59 00:17:09.839 ⇒ 00:17:15.568 Uttam Kumaran: actually like wrong with doing things this way from like a data side. See this.
60 00:17:19.079 ⇒ 00:17:19.679 Casie Aviles: Yes, yes.
61 00:17:19.925 ⇒ 00:17:21.399 Amber Lin: Yes, I can see your screen.
62 00:17:21.400 ⇒ 00:17:24.308 Uttam Kumaran: So what’s wrong here is
63 00:17:26.230 ⇒ 00:17:32.453 Uttam Kumaran: like, I can just take any of these right, for example. What is wrong here is,
64 00:17:34.950 ⇒ 00:17:41.330 Uttam Kumaran: you have sort of nested values mapped to single people.
65 00:17:43.650 ⇒ 00:17:51.409 Uttam Kumaran: And you have sort of text as well as
66 00:17:52.190 ⇒ 00:17:58.523 Uttam Kumaran: like you have. These are, I would consider, both text, but like this is like a different category versus this.
67 00:17:59.380 ⇒ 00:18:08.729 Uttam Kumaran: And in this row there is no inspector that can do. So you also have, like, input validation problems.
68 00:18:09.912 ⇒ 00:18:14.470 Uttam Kumaran: Right? So like another example of of this problem.
69 00:18:16.480 ⇒ 00:18:19.570 Amber Lin: They also have stuff on the top. Yeah.
70 00:18:19.570 ⇒ 00:18:24.049 Uttam Kumaran: So can you give me a sense of like how this is working today for them?
71 00:18:24.330 ⇒ 00:18:39.060 Amber Lin: So they would ask for a area. They understand what the area is they go into. Say, it’s in Corpus. They go into Corpus. They control F find the Zip code. It’s it was at the top that you were just in.
72 00:18:39.060 ⇒ 00:18:39.620 Uttam Kumaran: Yeah, yeah.
73 00:18:39.620 ⇒ 00:18:52.330 Amber Lin: And then they go check. Okay, zip code. It’s for rodent. Okay, it’s Larry. Oh, if it’s mosquito they look on top. Oh, it says everybody treats mosquito. They put Larry
74 00:18:52.510 ⇒ 00:18:57.030 Amber Lin: so it’s very. It’s very lots of extra steps.
75 00:18:57.430 ⇒ 00:19:02.239 Uttam Kumaran: So let’s take. Let’s let’s say, if we were gonna do that here we would. We would search a porpoise right.
76 00:19:02.240 ⇒ 00:19:09.000 Amber Lin: Yes, and then you identify the Zip code. Usually the Zip code is provided because the customer knows where they’re in.
77 00:19:09.000 ⇒ 00:19:09.670 Uttam Kumaran: So let’s say.
78 00:19:09.670 ⇒ 00:19:10.000 Amber Lin: That’s.
79 00:19:10.000 ⇒ 00:19:17.509 Uttam Kumaran: 7, 9, 2 corpus, and we want to find mosquito. It would be.
80 00:19:18.780 ⇒ 00:19:20.020 Amber Lin: Mosquitoes, on the on the.
81 00:19:20.020 ⇒ 00:19:21.429 Uttam Kumaran: We can do this right.
82 00:19:21.430 ⇒ 00:19:22.530 Amber Lin: Yes, yes, we will.
83 00:19:22.530 ⇒ 00:19:25.370 Uttam Kumaran: There isn’t anyone that does these right.
84 00:19:27.243 ⇒ 00:19:28.670 Uttam Kumaran: That’s correct.
85 00:19:30.510 ⇒ 00:19:33.149 Uttam Kumaran: So like, like, if I was gonna validate.
86 00:19:33.740 ⇒ 00:19:37.149 Uttam Kumaran: 7, 9, 8, 2. Let’s go to Corpus.
87 00:19:38.830 ⇒ 00:19:43.200 Uttam Kumaran: and let’s do the search for 7, 7, 9, 8, 2.
88 00:19:43.970 ⇒ 00:19:45.199 Uttam Kumaran: Yeah, it’s not here.
89 00:19:45.380 ⇒ 00:19:46.180 Amber Lin: Yes.
90 00:19:47.070 ⇒ 00:19:51.210 Uttam Kumaran: Cool, but then where? At 7, 7, 92.
91 00:19:51.210 ⇒ 00:20:00.349 Amber Lin: Might not be service. So that’s another. So that’s the next step that I want to do. So. They have this. If you scroll at the tab scroll, the tab. So the very last.
92 00:20:00.350 ⇒ 00:20:03.952 Uttam Kumaran: I’m just gonna leave one quick comment here. Just so we keep this like
93 00:20:09.043 ⇒ 00:20:09.626 Uttam Kumaran: service.
94 00:20:13.060 ⇒ 00:20:14.919 Uttam Kumaran: And then scroll all the way to the top. Sorry.
95 00:20:15.353 ⇒ 00:20:23.580 Amber Lin: Sorry. So go into the list of tabs. So scroll to the very, very right end of the
96 00:20:23.740 ⇒ 00:20:31.539 Amber Lin: different tabs. Yeah, it’s called service areas, copy copy of service areas. Yes.
97 00:20:32.350 ⇒ 00:20:34.959 Amber Lin: So after the the one after that.
98 00:20:35.560 ⇒ 00:20:50.889 Amber Lin: So they have this sheet of, okay, what zip codes are service? For which thing this is a separate spreadsheet. But ultimately I want to combine these so, or to have AI pass through this 1st to say, Okay, is this service?
99 00:20:51.180 ⇒ 00:20:57.119 Amber Lin: Yes? Then go look at the inspector sheet sheet to say, Okay, who is servicing that.
100 00:20:57.120 ⇒ 00:21:01.927 Uttam Kumaran: So I think what you can do is one. This is a great, this is a great
101 00:21:03.430 ⇒ 00:21:06.729 Uttam Kumaran: spreadsheet, like I think just we should.
102 00:21:07.690 ⇒ 00:21:11.880 Uttam Kumaran: I think this is a great solution, I think, instead, what you can do is
103 00:21:12.558 ⇒ 00:21:19.199 Uttam Kumaran: if if you can’t find it if you can’t find it.
104 00:21:19.380 ⇒ 00:21:23.781 Uttam Kumaran: Then basically, either you should look in the other.
105 00:21:24.590 ⇒ 00:21:30.910 Amber Lin: Service area spreadsheet. Take the No from there. Know in these
106 00:21:31.750 ⇒ 00:21:35.977 Uttam Kumaran: The other thing I would probably do is we’re gonna have to make it.
107 00:21:38.650 ⇒ 00:21:47.199 Uttam Kumaran: I think this works for the for the AI, and actually do think that this also works to replace their existing stuff.
108 00:21:47.200 ⇒ 00:21:48.330 Amber Lin: I agree.
109 00:21:49.880 ⇒ 00:21:56.049 Uttam Kumaran: and there’s a couple of different things like, if some people just want to look at past, you can create a past view of this.
110 00:21:56.710 ⇒ 00:22:00.069 Uttam Kumaran: But I actually feel like this is pretty quick.
111 00:22:00.070 ⇒ 00:22:05.660 Amber Lin: Yes, cause they schedule more than just pest sometimes, for in inspections.
112 00:22:05.850 ⇒ 00:22:09.830 Uttam Kumaran: I think my only suggestion here would be
113 00:22:33.110 ⇒ 00:22:41.589 Amber Lin: Or any potential risks and pitfalls or errors that you foresee might appear now or in the future.
114 00:22:43.290 ⇒ 00:22:45.381 Uttam Kumaran: Yeah sort of my only
115 00:22:53.780 ⇒ 00:22:57.489 Uttam Kumaran: sort of my only feedback is because
116 00:22:57.940 ⇒ 00:23:02.210 Uttam Kumaran: all of these folks all do. Each of these
117 00:23:02.330 ⇒ 00:23:06.980 Uttam Kumaran: you may be able to get a simpler table with just person, name.
118 00:23:16.790 ⇒ 00:23:26.316 Amber Lin: That is true. Some of them are general technicians. My only concern is that sometimes they have this specific person.
119 00:23:27.420 ⇒ 00:23:31.769 Amber Lin: who, like only does specialty lights.
120 00:23:31.990 ⇒ 00:23:39.360 Amber Lin: they only do lights, and then for lights. That then I have to add that specific specific person. So I rather just
121 00:23:39.490 ⇒ 00:23:44.520 Amber Lin: have it under different services and say, Okay, these are the people who treat it.
122 00:23:45.120 ⇒ 00:23:46.030 Uttam Kumaran: Hmm!
123 00:23:47.180 ⇒ 00:23:49.079 Uttam Kumaran: Give me one sec. Let me think about that.
124 00:24:41.390 ⇒ 00:24:44.400 Uttam Kumaran: Yeah, I feel like, the only other thing is.
125 00:24:44.890 ⇒ 00:24:48.300 Uttam Kumaran: you may want a list of all the technicians.
126 00:24:49.820 ⇒ 00:24:51.939 Uttam Kumaran: In one table, also.
127 00:24:51.940 ⇒ 00:24:56.419 Amber Lin: Hmm, I see. Yeah, cause the names are not standardized. That’s a good point.
128 00:24:57.740 ⇒ 00:25:00.615 Uttam Kumaran: The names are not standardized.
129 00:25:07.950 ⇒ 00:25:11.249 Amber Lin: And maybe to do a drop down menu instead.
130 00:25:12.034 ⇒ 00:25:28.319 Amber Lin: I do have all their names, and they do have all the technician names in a separate table. I think that will make them make the updating of this a lot easier as well. I think we can just make them drop down menus.
131 00:25:29.240 ⇒ 00:25:32.537 Uttam Kumaran: I honestly think, though like one thing, that
132 00:25:33.960 ⇒ 00:25:42.000 Uttam Kumaran: This is something that Casey you should test is, I wonder if you should
133 00:25:43.900 ⇒ 00:25:49.799 Uttam Kumaran: structure a super base request to get this versus getting just pasting this whole thing in there.
134 00:25:49.990 ⇒ 00:25:54.900 Uttam Kumaran: Reason being like someone says, Give me like.
135 00:25:55.190 ⇒ 00:26:03.120 Uttam Kumaran: like, basically you wanna you you need to know this potentially this, this
136 00:26:04.380 ⇒ 00:26:10.724 Uttam Kumaran: and then this right to to narrow it down, so it may be.
137 00:26:17.970 ⇒ 00:26:18.670 Casie Aviles: Yeah, make it.
138 00:26:18.670 ⇒ 00:26:25.460 Uttam Kumaran: It’d be better space, or at least query for like, give me everything in past first, st
139 00:26:26.800 ⇒ 00:26:30.699 Uttam Kumaran: if it takes too long. But you can try both ways. I just
140 00:26:33.420 ⇒ 00:26:35.700 Casie Aviles: Okay, yeah, I can do a spike on that.
141 00:26:36.320 ⇒ 00:26:42.299 Uttam Kumaran: I think the only the the the only concern I have is like, I wonder how often they’re changing these?
142 00:26:45.650 ⇒ 00:26:52.729 Uttam Kumaran: But I mean, for example, Amber, I guess my question would be, let’s say they want to remove Hannah.
143 00:26:53.600 ⇒ 00:26:55.700 Uttam Kumaran: Or like Hannah, is no longer working with the company.
144 00:26:57.190 ⇒ 00:27:00.670 Uttam Kumaran: That type of change, I think, is really hard to make in the system.
145 00:27:03.695 ⇒ 00:27:05.270 Amber Lin: could we? If
146 00:27:05.390 ⇒ 00:27:16.539 Amber Lin: if say, one person is no longer working, it’ll, be easier to do a find a replace all that will work, but if it’s moving them to a different
147 00:27:17.746 ⇒ 00:27:24.609 Amber Lin: different service, then they will have to filter for all of it, and then
148 00:27:25.140 ⇒ 00:27:28.750 Amber Lin: do it for different filters.
149 00:27:28.750 ⇒ 00:27:32.429 Uttam Kumaran: Wouldn’t it be better? Wouldn’t it be better to have a.
150 00:27:40.280 ⇒ 00:27:44.439 Uttam Kumaran: Cause. The zips are matched to the branch right.
151 00:27:44.780 ⇒ 00:27:45.530 Amber Lin: Yes.
152 00:27:45.940 ⇒ 00:27:48.770 Uttam Kumaran: So you don’t necessarily need
153 00:27:53.140 ⇒ 00:27:58.777 Uttam Kumaran: to blow up this by zips at this point, because I think you can do a
154 00:28:17.131 ⇒ 00:28:22.599 Amber Lin: You can. You can just do. I think it was, Tran. I did transpose and split
155 00:28:23.310 ⇒ 00:28:37.050 Amber Lin: together. But I think the thing is, say, for a specific branch if you scroll up a little bit for Wdis, not all of the inspectors service that.
156 00:28:38.430 ⇒ 00:28:42.710 Amber Lin: Then 4 branches, I think sometimes they also.
157 00:28:43.810 ⇒ 00:28:51.930 Uttam Kumaran: Let’s see, is this a proper? This will be a property of the person right.
158 00:28:53.040 ⇒ 00:28:54.960 Amber Lin: That is, that is true, that is.
159 00:28:54.960 ⇒ 00:28:56.240 Uttam Kumaran: What is Wdi? What is that?
160 00:28:56.300 ⇒ 00:29:00.240 Amber Lin: It’s like a wood damaging insect.
161 00:29:00.420 ⇒ 00:29:05.820 Amber Lin: It’s a specific inspection that they can do or can’t do.
162 00:29:07.680 ⇒ 00:29:12.049 Uttam Kumaran: So that seems to me that this is like a property of the person.
163 00:29:14.770 ⇒ 00:29:22.080 Uttam Kumaran: And then ideally, you have person Wdi and the branch.
164 00:29:22.680 ⇒ 00:29:25.409 Uttam Kumaran: and then you have another table that’s branched to zip.
165 00:29:27.530 ⇒ 00:29:30.200 Uttam Kumaran: like all the Zips in a in a branch right.
166 00:29:30.701 ⇒ 00:29:37.660 Amber Lin: Actually, if you go, can you go to Austin? Residential? I watch. I just want to check, because some of them.
167 00:29:37.840 ⇒ 00:29:40.569 Amber Lin: the branches not are not complete.
168 00:29:40.870 ⇒ 00:29:45.259 Amber Lin: They have some parts of it, but not all parts of it.
169 00:29:45.550 ⇒ 00:29:50.664 Amber Lin: Say some. Sometimes it’s just Austin North, and then sometimes there’s
170 00:29:51.540 ⇒ 00:29:59.410 Amber Lin: like it covers both. I just don’t know how complete it is, and if there’s different overlaps, then we might get issues.
171 00:29:59.410 ⇒ 00:30:00.747 Uttam Kumaran: I see. I see. I see
172 00:30:01.780 ⇒ 00:30:04.260 Uttam Kumaran: the way I’m thinking about it is like you have
173 00:30:04.860 ⇒ 00:30:08.559 Uttam Kumaran: one, Rosh. We have one sheet, that’s all. The inspectors.
174 00:30:09.630 ⇒ 00:30:13.900 Uttam Kumaran: Whether they’re wdi like other like
175 00:30:14.190 ⇒ 00:30:17.929 Uttam Kumaran: any other sort of things that they service right? So like
176 00:30:25.990 ⇒ 00:30:30.250 Uttam Kumaran: I guess it’s tough, because you have all of these specialties.
177 00:30:30.250 ⇒ 00:30:31.880 Amber Lin: Yes. Yeah.
178 00:30:32.040 ⇒ 00:30:40.080 Uttam Kumaran: So you need, you need basically this for every single person which it’s fine.
179 00:30:40.080 ⇒ 00:30:44.779 Amber Lin: Yeah, at that point, like I might as well map it to Zip, you know.
180 00:30:45.140 ⇒ 00:30:49.760 Uttam Kumaran: But that’s the thing like I think the Zip, mapping blows up, blows it up one more way.
181 00:30:55.560 ⇒ 00:31:02.480 Uttam Kumaran: I guess this is hard, because, see, like in some areas, you have.
182 00:31:04.337 ⇒ 00:31:07.199 Uttam Kumaran: I guess the dimensionality here is Zip. So
183 00:31:10.000 ⇒ 00:31:13.580 Uttam Kumaran: but then why is this? And like this person and this person.
184 00:31:13.880 ⇒ 00:31:18.200 Amber Lin: Oh, cause we copied that from the from the spreadsheet we can clean that up.
185 00:31:18.920 ⇒ 00:31:21.089 Uttam Kumaran: So if I go to college station.
186 00:31:22.520 ⇒ 00:31:24.139 Uttam Kumaran: Oh, I see.
187 00:31:24.140 ⇒ 00:31:37.209 Amber Lin: Yeah, that’s all we can clean that up in the different spreadsheets I was just concerned of, like, how accurate this would be. And, as you said, like, how updating this would work. But I just didn’t see a way of that.
188 00:31:37.210 ⇒ 00:31:39.710 Uttam Kumaran: I don’t know. Like to be to be honest.
189 00:31:41.200 ⇒ 00:31:44.060 Uttam Kumaran: this is only 600 rows.
190 00:31:44.740 ⇒ 00:31:48.855 Uttam Kumaran: It’s pretty easy for them to do like. For example, if
191 00:31:49.810 ⇒ 00:31:50.880 Amber Lin: You are making.
192 00:31:50.880 ⇒ 00:31:53.210 Amber Lin: Can I filter this little break?
193 00:31:53.750 ⇒ 00:31:59.909 Amber Lin: If you if it breaks we’ll revert it. It’s okay. You can. You can make a copy of it if you want.
194 00:32:03.090 ⇒ 00:32:04.989 Uttam Kumaran: Like if I was to. Just.
195 00:32:07.230 ⇒ 00:32:11.630 Uttam Kumaran: I guess mainly what I’m saying is like if someone was to go change something here.
196 00:32:11.630 ⇒ 00:32:12.799 Amber Lin: Yeah, yeah, I get it.
197 00:32:12.800 ⇒ 00:32:18.470 Uttam Kumaran: You know all they. But I guess my thing is like they can just change it directly in here, like you can almost.
198 00:32:19.610 ⇒ 00:32:23.250 Amber Lin: Yeah, I can do a group by group by branch.
199 00:32:23.410 ⇒ 00:32:33.058 Amber Lin: And then, yeah, I kept these so that we can valid. We can make sure that things are accurate in the Master Spreadsheet, because all of them are still be. Look up. I still want to do a last round.
200 00:32:33.300 ⇒ 00:32:35.719 Uttam Kumaran: Yeah. And then you can just copy paste the values.
201 00:32:35.720 ⇒ 00:32:40.570 Amber Lin: Yeah. And then I wanna delete like, Please.
202 00:32:40.570 ⇒ 00:32:47.669 Uttam Kumaran: It’s not like at that much. It’s only 600 sips versus like every zip in the country.
203 00:32:48.190 ⇒ 00:32:53.220 Uttam Kumaran: So it’s not that bad for them to go in here and be like cool. We’re adding a technician.
204 00:32:55.470 ⇒ 00:33:02.050 Uttam Kumaran: Let’s we’re adding a technician in the San Antonio branch who does lawn.
205 00:33:03.450 ⇒ 00:33:05.600 Uttam Kumaran: My biggest point would be
206 00:33:07.680 ⇒ 00:33:14.589 Uttam Kumaran: like if they were doing it here they would have to go. Add it to this like string, and then drag it down right.
207 00:33:14.590 ⇒ 00:33:16.249 Amber Lin: Yeah, I don’t think that’s too hard.
208 00:33:16.250 ⇒ 00:33:17.560 Uttam Kumaran: Okay, alright fine.
209 00:33:18.530 ⇒ 00:33:24.070 Uttam Kumaran: I mean, I don’t mind. Look, it’s better. It’s definitely better. I think there is an ideal state. But I kind of like. It’s a little bit.
210 00:33:24.070 ⇒ 00:33:29.389 Amber Lin: Yeah, it’s a little bit better. I think there’s still like this could be a lot
211 00:33:29.390 ⇒ 00:33:30.360 Amber Lin: great. I mean, you can.
212 00:33:30.360 ⇒ 00:33:31.630 Uttam Kumaran: But great. Yeah.
213 00:33:33.400 ⇒ 00:33:42.799 Amber Lin: Okay, great. Let’s, I think the next step will just to be validate if things are actually accurate, and then we can test what way we can do this with AI.
214 00:33:43.200 ⇒ 00:33:44.080 Uttam Kumaran: Okay.
215 00:33:44.270 ⇒ 00:33:47.699 Amber Lin: Yeah, yeah, I think the I think the biggest question I have is like.
216 00:33:49.019 ⇒ 00:33:52.969 Uttam Kumaran: One. We should add some of these to the Eval sheet.
217 00:33:53.760 ⇒ 00:33:57.770 Uttam Kumaran: Somebody’s like, Tell me, who’s the person in here? And it should give you the answer. Second.
218 00:33:58.000 ⇒ 00:34:00.117 Uttam Kumaran: yeah, I’m interested to see like whether
219 00:34:01.230 ⇒ 00:34:06.538 Uttam Kumaran: it does the right job. If you just paste this whole thing in, or if you have to actually do, look up.
220 00:34:10.300 ⇒ 00:34:14.879 Uttam Kumaran: Like if you need to go hit super base. Say, run a query for like
221 00:34:19.679 ⇒ 00:34:25.180 Uttam Kumaran: I mean, this table is not gonna fit exactly in super base pretty easily.
222 00:34:25.610 ⇒ 00:34:27.692 Uttam Kumaran: What you’ll have to do is
223 00:34:29.699 ⇒ 00:34:32.239 Amber Lin: I could flatten the rows from one.
224 00:34:32.239 ⇒ 00:34:32.979 Uttam Kumaran: Yeah.
225 00:34:32.980 ⇒ 00:34:43.230 Amber Lin: To just combine it in the title of that. Say, make it into a table column header. So then, when we put it in super base. It’s just okay. Look for the.
226 00:34:44.177 ⇒ 00:34:48.150 Uttam Kumaran: You add department, you should add department
227 00:34:48.770 ⇒ 00:34:51.850 Uttam Kumaran: commercial, residential, and this is columns.
228 00:34:55.000 ⇒ 00:34:56.575 Amber Lin: So what I mean.
229 00:34:57.780 ⇒ 00:34:59.660 Amber Lin: Sorry again. Please.
230 00:34:59.660 ⇒ 00:35:02.410 Uttam Kumaran: Like you should add department as a column.
231 00:35:05.590 ⇒ 00:35:08.779 Casie Aviles: The value would be either commercial or residential.
232 00:35:09.870 ⇒ 00:35:21.450 Uttam Kumaran: Well, basically, it’s it’s almost like you, you. If you’re gonna move it to super base, then you don’t care how long it is. So you would move department, commercial, generalized, and
233 00:35:21.670 ⇒ 00:35:24.559 Uttam Kumaran: whatever this is, if this is like the service.
234 00:35:24.800 ⇒ 00:35:28.820 Uttam Kumaran: all as columns, and then the last value is the this.
235 00:35:29.290 ⇒ 00:35:31.670 Uttam Kumaran: So you go from a really wide table
236 00:35:31.830 ⇒ 00:35:33.950 Uttam Kumaran: to a narrow but longer table.
237 00:35:34.490 ⇒ 00:35:35.040 Amber Lin: Oh!
238 00:35:35.040 ⇒ 00:35:36.070 Uttam Kumaran: You know what I mean?
239 00:35:37.970 ⇒ 00:35:40.980 Amber Lin: No, Casey, do you see it? Okay? As long as Casey sees it, it’s.
240 00:35:40.980 ⇒ 00:35:50.610 Uttam Kumaran: Meaning like depart department would go here. Commercial, go here, residential, go here. This service would go next, and then the value is the inspectors.
241 00:35:50.610 ⇒ 00:35:51.840 Amber Lin: Oh!
242 00:35:51.840 ⇒ 00:35:58.969 Uttam Kumaran: That. So cause because basically what you’re doing when you look at the sheet, you are saying cool. Let me do a pest salon. 1st column.
243 00:35:59.090 ⇒ 00:36:04.420 Uttam Kumaran: commercial non specialty mowing.
244 00:36:04.920 ⇒ 00:36:09.450 Uttam Kumaran: And then you’re doing Zip town branch, and then you get to the thing you want right.
245 00:36:09.450 ⇒ 00:36:10.160 Amber Lin: Yes.
246 00:36:10.160 ⇒ 00:36:15.910 Uttam Kumaran: So you’re going? 1, 2, 3, 4, 5, 6, 7 done right.
247 00:36:16.060 ⇒ 00:36:23.879 Amber Lin: Yeah, I mean, we. We can also, just usually people query for zip code. So we can get the row for all of it.
248 00:36:23.880 ⇒ 00:36:26.869 Uttam Kumaran: This is this is this is what they call like a cube.
249 00:36:27.450 ⇒ 00:36:29.570 Uttam Kumaran: They call this like a data cube.
250 00:36:30.470 ⇒ 00:36:32.290 Uttam Kumaran: They have, like the word for this.
251 00:36:32.590 ⇒ 00:36:32.980 Amber Lin: Oh!
252 00:36:32.980 ⇒ 00:36:38.660 Uttam Kumaran: Have dimensionality on multiple sides like you have dimensionality up here.
253 00:36:38.770 ⇒ 00:36:43.350 Uttam Kumaran: We have here. Typically the way to do this is just have
254 00:36:43.540 ⇒ 00:36:47.080 Uttam Kumaran: dimensionality on one side. But what you’re going to get is.
255 00:36:47.270 ⇒ 00:36:48.650 Amber Lin: A very flat.
256 00:36:48.650 ⇒ 00:36:51.930 Uttam Kumaran: Get a much longer table, but in super base it doesn’t matter.
257 00:36:52.090 ⇒ 00:36:52.900 Amber Lin: Okay.
258 00:36:52.900 ⇒ 00:36:58.240 Uttam Kumaran: Because in super base all, all Casey is gonna do is
259 00:36:58.650 ⇒ 00:37:03.780 Uttam Kumaran: take the input. So someone’s gonna say, cool. I want to know who the commercial
260 00:37:04.617 ⇒ 00:37:12.419 Uttam Kumaran: the commercial like, let’s say the commercial residential tree specialists are in 7, 8,
261 00:37:12.580 ⇒ 00:37:14.370 Uttam Kumaran: 7, 7, 8 0, 2.
262 00:37:15.090 ⇒ 00:37:18.659 Uttam Kumaran: And the query that’s gonna get structured is, gonna say, select
263 00:37:19.520 ⇒ 00:37:24.420 Uttam Kumaran: select technicians where zip code is 7, 7, 8 0, 2
264 00:37:24.540 ⇒ 00:37:30.939 Uttam Kumaran: department is this commercial is. This generalizes this services, this, and then it’ll hit in one value back.
265 00:37:31.530 ⇒ 00:37:32.810 Amber Lin: I see. Okay.
266 00:37:32.810 ⇒ 00:37:40.490 Uttam Kumaran: So it’s like it’s it’s easier to traverse for the database harder to traverse for us because we have to scroll.
267 00:37:40.810 ⇒ 00:37:43.309 Amber Lin: I see. Okay? I mean.
268 00:37:43.607 ⇒ 00:37:48.070 Uttam Kumaran: Think you can. You can. You can form that from this, so I wouldn’t change.
269 00:37:48.070 ⇒ 00:37:56.969 Amber Lin: Yeah, yeah, I agree. I think we can just test it if it just directly works. If we dump it into the Llm. And then, if not, we’ll transpose.
270 00:37:56.970 ⇒ 00:37:59.740 Uttam Kumaran: Yeah, you can then create a dynamic transpose on top of this.
271 00:38:00.660 ⇒ 00:38:02.489 Uttam Kumaran: And then ship that into super base.
272 00:38:02.490 ⇒ 00:38:04.399 Amber Lin: Okay, okay, that’s great.
273 00:38:04.400 ⇒ 00:38:12.598 Uttam Kumaran: Or you can even do that. All. Yeah, you can do whatever. But like that, if you’re gonna move the suit base, I would do it other. This isn’t this way. It’s gonna be
274 00:38:12.840 ⇒ 00:38:13.850 Amber Lin: Yeah, okay.
275 00:38:13.850 ⇒ 00:38:19.600 Uttam Kumaran: Be harder to traverse because you’re gonna have to dynamically do some columns. I wouldn’t be able to.
276 00:38:19.600 ⇒ 00:38:19.980 Amber Lin: Oh!
277 00:38:19.980 ⇒ 00:38:24.575 Uttam Kumaran: I would have it all as columns, and then the value is just this.
278 00:38:25.200 ⇒ 00:38:25.770 Amber Lin: Okay.
279 00:38:26.160 ⇒ 00:38:27.329 Uttam Kumaran: Yeah, that’s perfect.
280 00:38:27.330 ⇒ 00:38:30.269 Amber Lin: Great. Okay, I’m glad we called you. That’s really helpful.
281 00:38:31.670 ⇒ 00:38:35.955 Uttam Kumaran: Yeah, look up. If you, if you have interest, look up.
282 00:38:37.820 ⇒ 00:38:38.140 Amber Lin: Okay.
283 00:38:38.150 ⇒ 00:38:38.670 Amber Lin: Oh.
284 00:38:38.670 ⇒ 00:38:39.770 Uttam Kumaran: Go look that up.
285 00:38:39.770 ⇒ 00:38:43.699 Uttam Kumaran: It’s like, they say, a multi dimensional. See what you see, what I mean.
286 00:38:44.130 ⇒ 00:38:44.780 Amber Lin: Oh!
287 00:38:44.780 ⇒ 00:38:50.159 Uttam Kumaran: So you’re kind of like, cool. I wanted. I want, I want like Q 3, Asia.
288 00:38:50.450 ⇒ 00:38:54.700 Uttam Kumaran: And I want like Tvs. And so you can get that like queue right? So similarly.
289 00:38:55.130 ⇒ 00:38:55.560 Amber Lin: Yeah.
290 00:38:55.560 ⇒ 00:39:08.049 Uttam Kumaran: In your platform. You’re doing like cool. I want the one on one angle. There’s the department. There’s like the zip, and then there’s like another one. So you’re kind of so basically, what a what a what you’re doing is you’re taking this. And you’re just like.
291 00:39:09.890 ⇒ 00:39:12.700 Uttam Kumaran: I guess the way to explain it is you’re almost like
292 00:39:15.080 ⇒ 00:39:23.679 Uttam Kumaran: pushing. It’s almost like you. Then, instead of having 3 different dimensions, go get it. Each of these, it becomes just 3 columns, and every value so.
293 00:39:23.680 ⇒ 00:39:25.860 Amber Lin: Yeah, yeah, I I get what you mean.
294 00:39:25.860 ⇒ 00:39:34.170 Uttam Kumaran: I don’t. I don’t. I’m not like a visual person, so like I wasn’t good at calculus. I can’t really explain how this like shit the shape shifts. But you know what I mean. Yeah.
295 00:39:34.170 ⇒ 00:39:35.140 Amber Lin: Yeah, I know.
296 00:39:35.310 ⇒ 00:39:40.020 Uttam Kumaran: Okay. But olap cube is like what the name of this thing is.
297 00:39:40.020 ⇒ 00:39:40.730 Amber Lin: Okay.
298 00:39:41.960 ⇒ 00:39:43.080 Uttam Kumaran: Awesome.
299 00:39:44.950 ⇒ 00:39:46.370 Amber Lin: Great! I guess.
300 00:39:46.550 ⇒ 00:39:57.870 Amber Lin: Casey, we can do a quick test of if it works with. If we just dump it in. If not, we’ll look at how to flatten it together, and then I’ll make a ticket for that.
301 00:39:59.010 ⇒ 00:40:00.149 Casie Aviles: Okay. Yeah.
302 00:40:00.150 ⇒ 00:40:01.750 Amber Lin: All right. Thank you, Tom.
303 00:40:03.270 ⇒ 00:40:05.060 Uttam Kumaran: Thank you, appreciate it.
304 00:40:05.060 ⇒ 00:40:06.050 Amber Lin: Alrighty!
305 00:40:06.470 ⇒ 00:40:07.500 Amber Lin: Bye-bye.
306 00:40:07.500 ⇒ 00:40:08.260 Uttam Kumaran: Bye.