Meeting Title: Brainforge x ABC Migration and Zip Codes Date: 2026-02-25 Meeting participants: Casie Aviles, Pranav Narahari, Samuel Roberts, Mustafa Raja, Amber Lin
WEBVTT
1 00:01:21.890 ⇒ 00:01:22.590 Mustafa Raja: Wait.
2 00:01:28.970 ⇒ 00:01:30.090 Mustafa Raja: Hey, can you hear me?
3 00:01:32.690 ⇒ 00:01:33.670 Casie Aviles: Hey, yeah.
4 00:01:34.690 ⇒ 00:01:35.440 Samuel Roberts: Hi, Mike.
5 00:02:06.690 ⇒ 00:02:07.930 Amber Lin: Hi, team.
6 00:02:11.880 ⇒ 00:02:13.050 Amber Lin: Boom.
7 00:02:13.420 ⇒ 00:02:22.319 Amber Lin: Let’s see, we have… Some transfer stuff, which, like, the stick is in my hand, and then…
8 00:02:23.110 ⇒ 00:02:29.159 Amber Lin: We have the central doc stuff, zip codes, and migration.
9 00:02:29.410 ⇒ 00:02:36.520 Amber Lin: So… I… Which one is the smallest that we can talk about first?
10 00:02:41.700 ⇒ 00:02:42.410 Samuel Roberts: I’m enough.
11 00:02:43.120 ⇒ 00:02:46.450 Casie Aviles: The zip code stuff is pretty big, I’d say.
12 00:02:46.450 ⇒ 00:02:47.219 Mustafa Raja: No, I…
13 00:02:47.220 ⇒ 00:02:48.210 Amber Lin: Oh, okay.
14 00:02:48.210 ⇒ 00:02:50.879 Mustafa Raja: Yeah, I would also want to talk about the…
15 00:02:51.650 ⇒ 00:02:54.980 Mustafa Raja: a migration proposal for AI models?
16 00:02:54.980 ⇒ 00:02:55.670 Samuel Roberts: Yes.
17 00:02:55.670 ⇒ 00:02:59.290 Amber Lin: Okay. Let’s talk about that, then.
18 00:03:00.050 ⇒ 00:03:03.010 Mustafa Raja: So, let’s talk about migrating.
19 00:03:03.650 ⇒ 00:03:04.999 Mustafa Raja: Let me share my screen, then.
20 00:03:05.730 ⇒ 00:03:06.390 Samuel Roberts: Cool.
21 00:03:11.050 ⇒ 00:03:15.990 Mustafa Raja: Okay… put this… this, okay.
22 00:03:16.150 ⇒ 00:03:18.919 Mustafa Raja: Yeah, so…
23 00:03:19.890 ⇒ 00:03:31.560 Mustafa Raja: So we have two ways to connect with, Google, right? One is Vertex API, and then one is Gemini API. Vertex, API, the benefit,
24 00:03:31.580 ⇒ 00:03:48.419 Mustafa Raja: whether it would be, you know, we might be able to, integrate it using the service accounts or something, and this is just plain old, API key that’s giving us access. Now, I don’t know, how Vertex CI would play with…
25 00:03:48.420 ⇒ 00:03:49.389 Samuel Roberts: Oh, yeah. Huh?
26 00:03:50.300 ⇒ 00:03:55.410 Mustafa Raja: If… if… If you know it’s, it’s working with these.
27 00:03:57.490 ⇒ 00:03:58.150 Samuel Roberts: Right.
28 00:03:58.910 ⇒ 00:04:06.049 Mustafa Raja: But, yeah, I just need to know, you know, what the team feels about, which tool would team feel good about.
29 00:04:06.610 ⇒ 00:04:09.550 Mustafa Raja: using Vertex AI or Gemini API?
30 00:04:11.940 ⇒ 00:04:16.430 Samuel Roberts: To be honest, I don’t know much about the Vertex API, I just know people have complained about Mike.
31 00:04:17.040 ⇒ 00:04:25.669 Samuel Roberts: getting access to it, and Google doesn’t make it easy. I don’t know, I’m looking now just to see… I can dig a little more…
32 00:04:25.670 ⇒ 00:04:30.059 Mustafa Raja: Before committing to any, I just wanted to talk about that.
33 00:04:30.060 ⇒ 00:04:30.590 Samuel Roberts: Yeah.
34 00:04:30.590 ⇒ 00:04:40.240 Mustafa Raja: models themselves, I think we should be looking forward to 3.1 Pro and 3.1 Flash, right? Or 3 Flash, I believe. It’s 3 Flash.
35 00:04:40.240 ⇒ 00:04:43.729 Samuel Roberts: Yeah, I think it’s 3 flash, but yes, I think that’s probably where we want to be.
36 00:04:43.730 ⇒ 00:04:45.679 Mustafa Raja: Okay, and we want both of those, right?
37 00:04:45.820 ⇒ 00:04:47.080 Mustafa Raja: Just making sure.
38 00:04:47.550 ⇒ 00:04:53.310 Samuel Roberts: I think so, I was looking at it a little bit this morning, I, I think…
39 00:04:54.280 ⇒ 00:05:01.390 Samuel Roberts: I think we can probably make use of both of them for certain things, in order to get the kind of same thing we were getting out of 4.0.
40 00:05:01.760 ⇒ 00:05:04.319 Samuel Roberts: I just know it’s a little bit… you know.
41 00:05:05.180 ⇒ 00:05:08.810 Samuel Roberts: different. You know, we haven’t upgraded 4-0 in a while, so…
42 00:05:08.810 ⇒ 00:05:09.300 Mustafa Raja: Probably a lot.
43 00:05:10.050 ⇒ 00:05:10.970 Samuel Roberts: That could change.
44 00:05:13.440 ⇒ 00:05:19.900 Samuel Roberts: Okay. Yeah, let me do a little bit digging into Gemini versus Vertex. We may even have to get Tim’s…
45 00:05:20.850 ⇒ 00:05:24.139 Samuel Roberts: Thoughts on that, if they care about certain… what it says here.
46 00:05:24.360 ⇒ 00:05:29.840 Samuel Roberts: Most developers should use the Gemini Developer API unless there’s a need for specific enterprise controls, so I’m.
47 00:05:29.840 ⇒ 00:05:35.949 Mustafa Raja: Yeah, yeah, I think… let me see if I still have that page…
48 00:05:39.130 ⇒ 00:05:51.420 Mustafa Raja: Yeah, for Vertex AI, it’s advertising itself as, you know, building AI agents, integrating AI into your applications, whereas this is just an API directly to, what’s it called?
49 00:05:52.330 ⇒ 00:05:54.140 Mustafa Raja: The models itself, no?
50 00:05:54.370 ⇒ 00:05:55.290 Mustafa Raja: So this makes…
51 00:05:55.290 ⇒ 00:05:56.220 Samuel Roberts: Yeah.
52 00:05:56.240 ⇒ 00:06:03.060 Mustafa Raja: we don’t need this, we are working with Mastra. I just need… I just… I just want to communicate that.
53 00:06:03.620 ⇒ 00:06:10.199 Samuel Roberts: Yeah, I think… I think Gemini should be fine.
54 00:06:10.380 ⇒ 00:06:12.790 Samuel Roberts: This is what I was just looking at, that was similar.
55 00:06:13.390 ⇒ 00:06:14.500 Mustafa Raja: Oh, okay.
56 00:06:25.710 ⇒ 00:06:30.769 Samuel Roberts: So I think, yeah, I think we’re probably good with Gemini unless they specifically need something from Vertex, but I don’t…
57 00:06:32.020 ⇒ 00:06:34.540 Samuel Roberts: I think it’ll be… Needed.
58 00:06:37.050 ⇒ 00:06:39.419 Mustafa Raja: Yep, I’m just going to remove vertex then, right?
59 00:06:40.100 ⇒ 00:06:43.280 Samuel Roberts: Okay, yeah, unless Tim says we want that specifically.
60 00:06:43.750 ⇒ 00:06:48.190 Mustafa Raja: So that should still be in proposal, then? Put into the calendar.
61 00:06:49.050 ⇒ 00:06:55.759 Samuel Roberts: I might just mention it, but I don’t think I’d give it as a… yeah.
62 00:06:57.380 ⇒ 00:07:02.649 Samuel Roberts: I would leave that unless he thinks… you know what I mean? Unless he knows something about Vertex that we don’t, I think this is fine.
63 00:07:04.270 ⇒ 00:07:04.970 Samuel Roberts: Okay.
64 00:07:12.910 ⇒ 00:07:23.160 Mustafa Raja: Yeah, I might, reboot this, and then I might reshare it, reshare the notion after. Okay. Yeah, we are just proposing Gemini API, then.
65 00:07:23.500 ⇒ 00:07:25.600 Mustafa Raja: Anyhow, I’ll just rebirth all of this.
66 00:07:25.710 ⇒ 00:07:28.090 Mustafa Raja: Okay.
67 00:07:28.560 ⇒ 00:07:31.089 Mustafa Raja: That’s pretty much it for the proposal.
68 00:07:32.330 ⇒ 00:07:32.850 Samuel Roberts: Cool.
69 00:07:33.000 ⇒ 00:07:34.150 Samuel Roberts: I think that’s good then.
70 00:07:34.780 ⇒ 00:07:38.510 Samuel Roberts: We’ll have to do a little bit of testing once we get those up, but… we can take that out.
71 00:07:38.510 ⇒ 00:07:41.689 Mustafa Raja: Yeah, we’ll have to… we’ll have to adjust a lot of prompts.
72 00:07:42.340 ⇒ 00:07:42.940 Samuel Roberts: Yeah.
73 00:07:44.560 ⇒ 00:07:52.900 Mustafa Raja: I don’t know if they care about, how Andy responds, you know? If the, if the response style is a little different, if CSRs would care about that.
74 00:07:55.690 ⇒ 00:07:58.489 Samuel Roberts: That’s a good question, I imagine.
75 00:07:58.490 ⇒ 00:08:07.980 Amber Lin: I think they’re used to, Andy, giving slightly different responses. It’s just say, for example, certain fields that they need that, for example.
76 00:08:08.020 ⇒ 00:08:19.650 Amber Lin: I think when they asked for inspectors, and then we used to specify if it’s residential or commercial, and then we stopped specifying that, and then they wanted that back.
77 00:08:19.790 ⇒ 00:08:22.300 Amber Lin: So I think things like that are…
78 00:08:23.790 ⇒ 00:08:24.180 Mustafa Raja: Okay.
79 00:08:24.310 ⇒ 00:08:28.299 Amber Lin: What we include, not necessarily in the form items.
80 00:08:28.540 ⇒ 00:08:30.129 Samuel Roberts: Okay, I think we’ll be alright.
81 00:08:30.390 ⇒ 00:08:31.840 Samuel Roberts: Yeah, that’s fine.
82 00:08:36.740 ⇒ 00:08:37.490 Amber Lin: look at…
83 00:08:37.490 ⇒ 00:08:37.980 Samuel Roberts: Okay.
84 00:08:38.140 ⇒ 00:08:39.049 Amber Lin: this.
85 00:08:41.500 ⇒ 00:08:45.690 Amber Lin: Okay. Was these two done?
86 00:08:47.440 ⇒ 00:08:49.390 Mustafa Raja: Oh, by the way, was done.
87 00:08:50.230 ⇒ 00:08:51.110 Amber Lin: Okay.
88 00:08:51.250 ⇒ 00:08:54.669 Amber Lin: What about thumbs down the feedback logging?
89 00:08:56.880 ⇒ 00:09:04.319 Mustafa Raja: This was a throwing error that we fixed later in the week, so I couldn’t migrate it until then. Now I can migrate this.
90 00:09:04.650 ⇒ 00:09:05.300 Amber Lin: Okay.
91 00:09:05.300 ⇒ 00:09:09.890 Mustafa Raja: The workflow isn’t working. That’s why we saw a gap in dashboard also.
92 00:09:10.320 ⇒ 00:09:15.510 Amber Lin: Okay, gotcha. So, we will need to do that before we’ll have done any tests.
93 00:09:16.900 ⇒ 00:09:18.670 Amber Lin: That’s the new stage name.
94 00:09:18.890 ⇒ 00:09:19.530 Mustafa Raja: Yeah.
95 00:09:19.690 ⇒ 00:09:22.699 Amber Lin: So, let’s prioritize that and get that.
96 00:09:23.010 ⇒ 00:09:24.249 Amber Lin: Put that down.
97 00:09:24.350 ⇒ 00:09:27.300 Amber Lin: Alright.
98 00:09:27.540 ⇒ 00:09:31.139 Amber Lin: Is this… is this going to happen this week?
99 00:09:34.820 ⇒ 00:09:36.480 Amber Lin: Validate alerts.
100 00:09:39.570 ⇒ 00:09:43.009 Mustafa Raja: We already have Slack integrated.
101 00:09:44.030 ⇒ 00:09:46.730 Samuel Roberts: Yeah, I think it was just creating certain scenarios where we would…
102 00:09:46.920 ⇒ 00:09:51.639 Mustafa Raja: Yeah. Know that it’s working right. I could do that if we want to verify it again.
103 00:09:52.390 ⇒ 00:09:58.650 Samuel Roberts: I think it’s okay. I think I would kick the… I mean, we can do more failure testing, but I don’t think it’s as critical as some of this other stuff.
104 00:09:59.040 ⇒ 00:09:59.909 Mustafa Raja: Yeah. Okay.
105 00:10:01.250 ⇒ 00:10:03.970 Samuel Roberts: Like, we tested the Slack alerts briefly, so I think we’re good.
106 00:10:04.930 ⇒ 00:10:07.789 Amber Lin: Okay, I’m gonna put it here and say it’s time for…
107 00:10:08.330 ⇒ 00:10:11.590 Amber Lin: Okay, we have staging up, right?
108 00:10:13.670 ⇒ 00:10:15.559 Casie Aviles: No, no, it’s not yet up.
109 00:10:16.470 ⇒ 00:10:17.589 Amber Lin: Oh, okay.
110 00:10:18.000 ⇒ 00:10:22.690 Amber Lin: So these two are… this is in progress.
111 00:10:25.800 ⇒ 00:10:28.829 Casie Aviles: Yeah, we should, yeah, it should be. Okay.
112 00:10:28.830 ⇒ 00:10:38.459 Amber Lin: And I know Utam was talking about, having real points BigQuery, or something. Was that done, or should we…
113 00:10:38.460 ⇒ 00:10:44.590 Casie Aviles: Yeah, I think he worked on it already. It should… I was able to see it on Rel. He made, like, a new…
114 00:10:45.670 ⇒ 00:10:48.149 Casie Aviles: Explorer dashboard, I think, yeah.
115 00:10:48.770 ⇒ 00:10:52.990 Amber Lin: Oh, okay, so it points to… so I’ll say it points to BigQuery.
116 00:10:59.410 ⇒ 00:11:02.800 Amber Lin: And then this one, tables for…
117 00:11:04.000 ⇒ 00:11:07.360 Amber Lin: Is this something we’re doing this week, or later?
118 00:11:10.180 ⇒ 00:11:15.760 Casie Aviles: Tables for zip code. No, it’s not, it’s not yet, happening.
119 00:11:16.490 ⇒ 00:11:20.450 Samuel Roberts: Yeah, but yeah, I don’t think that’s… Yeah, it’s fine.
120 00:11:21.380 ⇒ 00:11:22.030 Amber Lin: Alright.
121 00:11:22.250 ⇒ 00:11:28.760 Amber Lin: And then… Is that everything on the migration side, other than, say, the zip codes?
122 00:11:31.990 ⇒ 00:11:35.730 Amber Lin: Like, what are we… what will stop us from migration so far?
123 00:11:35.730 ⇒ 00:11:36.490 Samuel Roberts: I think…
124 00:11:37.170 ⇒ 00:11:41.409 Samuel Roberts: If we’re gonna be switching over the models, we probably need to do some more testing there, then, part of this.
125 00:11:42.000 ⇒ 00:11:42.510 Amber Lin: Okay.
126 00:11:42.510 ⇒ 00:11:43.190 Casie Aviles: Just a log.
127 00:11:43.190 ⇒ 00:11:44.080 Amber Lin: 100%.
128 00:11:44.080 ⇒ 00:11:45.139 Samuel Roberts: This needs to get broken down.
129 00:11:45.970 ⇒ 00:11:50.130 Amber Lin: Yeah, I think let’s extend this, because we did a start grasp.
130 00:11:50.450 ⇒ 00:11:51.530 Samuel Roberts: Yeah, that’s fine.
131 00:11:52.800 ⇒ 00:11:53.540 Amber Lin: Cool.
132 00:11:53.690 ⇒ 00:11:58.469 Amber Lin: Alright, so if that’s all, could we get this?
133 00:11:58.870 ⇒ 00:12:03.980 Amber Lin: done and flushed out by end of this week? Like, when can I say, clients, you can test this?
134 00:12:09.790 ⇒ 00:12:11.860 Samuel Roberts: I mean… Excuse me.
135 00:12:12.320 ⇒ 00:12:19.340 Samuel Roberts: We could probably get something, and get some… yeah, as long as they know it’s not production, right?
136 00:12:19.960 ⇒ 00:12:24.699 Samuel Roberts: I’m… I think we’re… pretty good to do that, I mean, correct me if I’m wrong.
137 00:12:25.450 ⇒ 00:12:33.760 Samuel Roberts: Mustafa, or Casey, I’m not sure what else is stopping us from setting up a staging environment there.
138 00:12:34.750 ⇒ 00:12:36.659 Samuel Roberts: Except it’s on the old model still.
139 00:12:39.080 ⇒ 00:12:39.690 Casie Aviles: Yeah.
140 00:12:39.690 ⇒ 00:12:44.559 Mustafa Raja: I believe we wanted to, you know, integrate NHN with it.
141 00:12:44.960 ⇒ 00:12:47.500 Mustafa Raja: With the development environment that we have.
142 00:12:49.430 ⇒ 00:12:51.129 Samuel Roberts: Oh, for the.
143 00:12:51.520 ⇒ 00:12:53.889 Mustafa Raja: To see how it’s acting up, you know?
144 00:12:54.270 ⇒ 00:12:55.089 Samuel Roberts: That’s right, yeah.
145 00:12:55.090 ⇒ 00:12:56.010 Mustafa Raja: Questions?
146 00:12:56.550 ⇒ 00:12:58.380 Samuel Roberts: Right, we wanted to test that more, didn’t we?
147 00:13:00.250 ⇒ 00:13:01.790 Mustafa Raja: Can we add a week here?
148 00:13:02.420 ⇒ 00:13:05.530 Samuel Roberts: I mean, I think that could be this validation step, if we want it to be.
149 00:13:05.530 ⇒ 00:13:06.109 Mustafa Raja: Which one?
150 00:13:06.350 ⇒ 00:13:07.309 Mustafa Raja: The staging validity.
151 00:13:09.370 ⇒ 00:13:10.200 Samuel Roberts: Yeah.
152 00:13:10.780 ⇒ 00:13:11.300 Mustafa Raja: Hmm.
153 00:13:14.670 ⇒ 00:13:21.309 Mustafa Raja: Yeah, let’s keep these, and then maybe, maybe next week we can, I’ll ask them, you know.
154 00:13:22.040 ⇒ 00:13:22.500 Samuel Roberts: Yeah.
155 00:13:22.500 ⇒ 00:13:23.040 Amber Lin: Okay.
156 00:13:24.350 ⇒ 00:13:27.539 Amber Lin: That will be… that will be the goal, then.
157 00:13:28.210 ⇒ 00:13:32.480 Amber Lin: Let me move… S… Pop.
158 00:13:32.580 ⇒ 00:13:34.230 Amber Lin: Oh, here.
159 00:13:36.360 ⇒ 00:13:37.030 Amber Lin: Cool.
160 00:13:37.320 ⇒ 00:13:45.530 Amber Lin: Okay, if that’s all on migration, let’s, let’s make sure these are done, and then let’s talk about the zip code stuff.
161 00:13:46.830 ⇒ 00:13:50.670 Casie Aviles: Okay. Can I…
162 00:13:50.670 ⇒ 00:13:51.230 Amber Lin: Yeah, go ahead.
163 00:13:51.230 ⇒ 00:13:52.900 Casie Aviles: share my screen, okay.
164 00:13:58.130 ⇒ 00:13:58.940 Casie Aviles: Alright.
165 00:13:59.200 ⇒ 00:14:02.919 Casie Aviles: Yeah, you can see the screen now, guys, right?
166 00:14:06.020 ⇒ 00:14:06.750 Casie Aviles: Hello?
167 00:14:06.750 ⇒ 00:14:08.170 Samuel Roberts: Yes. Yes.
168 00:14:08.380 ⇒ 00:14:12.050 Casie Aviles: Okay, yeah, okay, so for, for the zip…
169 00:14:13.090 ⇒ 00:14:17.030 Casie Aviles: Yeah, for the zip code, so I’ve just been creating all these validation sheets.
170 00:14:18.410 ⇒ 00:14:24.350 Casie Aviles: I’ve been trying to… Resolve as much… discrepancy that I can find.
171 00:14:25.220 ⇒ 00:14:30.700 Casie Aviles: So, to give you an idea, or also just a recap of what I was trying to do.
172 00:14:31.180 ⇒ 00:14:37.730 Casie Aviles: Basically, we have all these multiple, spreadsheets that also has their own sheets inside.
173 00:14:39.890 ⇒ 00:14:43.530 Casie Aviles: For example, they’re not very…
174 00:14:44.970 ⇒ 00:14:49.530 Casie Aviles: They’re not the same, like, structure for all… for each, so…
175 00:14:50.120 ⇒ 00:14:54.370 Casie Aviles: what I did was I’ve been…
176 00:14:54.480 ⇒ 00:14:56.829 Casie Aviles: I’ve been using, like, a script
177 00:14:57.820 ⇒ 00:15:01.620 Casie Aviles: Or I’ve been creating scripts using Purser, so… hold on.
178 00:15:07.550 ⇒ 00:15:11.640 Casie Aviles: Yeah, so I’ve been using these scripts to… Python scripts to…
179 00:15:11.870 ⇒ 00:15:15.279 Casie Aviles: Create, create the normalized table, so…
180 00:15:16.200 ⇒ 00:15:22.360 Casie Aviles: Basically, I have several Python scripts over here, and… That’s what…
181 00:15:22.640 ⇒ 00:15:26.029 Casie Aviles: That’s what I use in order to write these normalized.
182 00:15:26.160 ⇒ 00:15:30.550 Casie Aviles: Sheets, so… So us…
183 00:15:31.260 ⇒ 00:15:33.649 Casie Aviles: So right now, what I can…
184 00:15:34.370 ⇒ 00:15:40.600 Casie Aviles: what I was able to, like, get was to have the records existing, or, like, at least the names.
185 00:15:42.240 ⇒ 00:15:48.710 Casie Aviles: But with the zip count match, it’s not, like, perfect, so… I’m…
186 00:15:48.980 ⇒ 00:15:56.570 Casie Aviles: Not 100% sure, like, we can guarantee, like, complete… a complete exact match with all of the sheets involved.
187 00:15:59.240 ⇒ 00:16:02.679 Samuel Roberts: Okay, so if this is zero, it’s green, it’s good? Is that what…
188 00:16:03.220 ⇒ 00:16:05.730 Casie Aviles: Yeah, that’s what it means, so right now.
189 00:16:05.730 ⇒ 00:16:07.240 Samuel Roberts: It’s matching? Okay.
190 00:16:07.240 ⇒ 00:16:15.050 Casie Aviles: Yeah, if it’s yellow or positive, it’s… that means that the table or our database has more records.
191 00:16:15.400 ⇒ 00:16:20.570 Casie Aviles: Compared to, like, What we have, what the source sheet has.
192 00:16:20.830 ⇒ 00:16:25.270 Casie Aviles: Okay. And then, if it’s a negative, then it means we’re missing, like, an assignment.
193 00:16:26.250 ⇒ 00:16:28.930 Casie Aviles: So, this is what we have for lawn.
194 00:16:30.380 ⇒ 00:16:35.019 Casie Aviles: And then, for example, this one, the source sheet didn’t have any zips.
195 00:16:35.130 ⇒ 00:16:38.950 Casie Aviles: written, so… Basically, what this means is I just…
196 00:16:39.860 ⇒ 00:16:44.450 Casie Aviles: Everything that has, like, an Austin branch, then that I would assign it there.
197 00:16:47.020 ⇒ 00:16:51.969 Casie Aviles: This one’s for mechanical, so… yeah, this was, like, an earlier version that was…
198 00:16:52.350 ⇒ 00:16:56.379 Casie Aviles: That was, so it’s not very… it’s not the same all throughout, but…
199 00:16:56.560 ⇒ 00:16:58.510 Casie Aviles: I have to customize it for each.
200 00:16:59.220 ⇒ 00:17:05.590 Casie Aviles: sheet… And then I also have, like, the inspector sheet over here. Oh, wait.
201 00:17:06.290 ⇒ 00:17:08.029 Casie Aviles: So this is just,
202 00:17:11.050 ⇒ 00:17:13.169 Casie Aviles: Yeah, this is how it looks like right now.
203 00:17:15.089 ⇒ 00:17:21.900 Casie Aviles: So I think we have, like, most of the names now, I’ve resolved that, but there are zip count matches that are still…
204 00:17:25.599 ⇒ 00:17:29.539 Casie Aviles: the Excel. So basically, that’s… what I have right now.
205 00:17:30.430 ⇒ 00:17:32.099 Casie Aviles: So I was wondering, like, what’s…
206 00:17:32.240 ⇒ 00:17:36.389 Casie Aviles: Kind of, like, the best way to do, like, another review of this, or, like.
207 00:17:38.690 ⇒ 00:17:47.000 Casie Aviles: Yeah, what’s, like, the best next step here? Because the next thing I wanted to work on was also, like, automated testing.
208 00:17:47.160 ⇒ 00:17:52.230 Casie Aviles: So we could, like, just… for example, we have these test questions, and then we’re just gonna run…
209 00:17:52.530 ⇒ 00:17:57.419 Casie Aviles: Through each of these, and then it’s gonna output, like, the responses here.
210 00:17:57.690 ⇒ 00:18:00.700 Casie Aviles: And then the SQL query that was generated.
211 00:18:02.600 ⇒ 00:18:06.549 Casie Aviles: That’s… that’s what I was going… yeah, I’m also planning to work on.
212 00:18:07.760 ⇒ 00:18:14.260 Samuel Roberts: That sounds good for testing. Can you go back to the normalized spreadsheet real quick? Or the tab? So, do we have a sense of,
213 00:18:15.010 ⇒ 00:18:19.720 Samuel Roberts: How many of these are… Red or yellow?
214 00:18:21.840 ⇒ 00:18:24.710 Samuel Roberts: I don’t have, like, a… Okay.
215 00:18:24.960 ⇒ 00:18:29.560 Casie Aviles: I don’t have, like, a percentage yet, but we can do a filter.
216 00:18:30.030 ⇒ 00:18:32.020 Samuel Roberts: Yeah, I was just curious.
217 00:18:43.490 ⇒ 00:18:45.110 Casie Aviles: Yeah, for example, it’s fine.
218 00:18:45.110 ⇒ 00:18:45.750 Samuel Roberts: Okay.
219 00:18:49.280 ⇒ 00:18:51.550 Samuel Roberts: Oh, and it’s not interesting, okay.
220 00:18:51.680 ⇒ 00:18:54.189 Casie Aviles: Yeah, that’s… yeah, that’s also because there are no…
221 00:18:55.150 ⇒ 00:18:57.699 Casie Aviles: Zips specified on the sheet, so…
222 00:18:59.940 ⇒ 00:19:02.320 Casie Aviles: There are no zip codes specified on this sheet.
223 00:19:02.570 ⇒ 00:19:03.040 Casie Aviles: So…
224 00:19:03.040 ⇒ 00:19:03.880 Samuel Roberts: Right, right.
225 00:19:03.880 ⇒ 00:19:06.259 Casie Aviles: Yeah. Yep. Has to be on occasions.
226 00:19:07.730 ⇒ 00:19:11.900 Samuel Roberts: Oh, that’s why it’s 600… okay, I was wondering why it was such a high number. Okay.
227 00:19:18.040 ⇒ 00:19:26.230 Samuel Roberts: Yeah, I’m just curious, I mean, I’m just trying to think if there’s a good way to resolve these while you’re doing the testing, maybe. Or getting the testing set up, at least.
228 00:19:27.750 ⇒ 00:19:33.430 Amber Lin: I’m wondering if this could be some stuff that’s not even…
229 00:19:33.660 ⇒ 00:19:38.500 Amber Lin: Does not even have zip codes. For example, can you go to the inspector sheet tabs?
230 00:19:40.770 ⇒ 00:19:42.269 Casie Aviles: This one? The source?
231 00:19:42.710 ⇒ 00:19:47.049 Amber Lin: Yeah, let’s see if… for example, I guess I’m…
232 00:19:47.450 ⇒ 00:20:01.040 Amber Lin: On… yeah, the drywood turbine fumigation lease in Austin goes to Jonathan Showbert. Like, that type of stuff, they have it in every single tab. How do we account for those right now?
233 00:20:02.470 ⇒ 00:20:04.320 Casie Aviles: Yeah, right now, these are just…
234 00:20:04.470 ⇒ 00:20:11.180 Casie Aviles: These are not… they don’t have, like, any zip codes in the normalized sheet, but we have, like.
235 00:20:11.460 ⇒ 00:20:13.020 Casie Aviles: this service area.
236 00:20:13.860 ⇒ 00:20:19.580 Samuel Roberts: Okay, so we should… Do we need to match that to every zip in that service area, then?
237 00:20:19.750 ⇒ 00:20:21.159 Samuel Roberts: Is that the idea here?
238 00:20:23.760 ⇒ 00:20:26.580 Casie Aviles: Yeah, I think, I think that’s what we can do, so…
239 00:20:27.060 ⇒ 00:20:34.180 Casie Aviles: Instead of, like, having zero zips here, we’re just gonna have, like, All the zip codes.
240 00:20:34.450 ⇒ 00:20:35.640 Samuel Roberts: Under that…
241 00:20:35.640 ⇒ 00:20:37.070 Casie Aviles: Branch, for example.
242 00:20:37.910 ⇒ 00:20:39.210 Samuel Roberts: Oh…
243 00:20:39.680 ⇒ 00:20:47.190 Amber Lin: I see. But I just want to confirm, like, those text, rows are also accounted for, right?
244 00:20:49.190 ⇒ 00:20:51.720 Casie Aviles: Yeah, I tried to account for that.
245 00:20:53.290 ⇒ 00:20:56.319 Casie Aviles: But then again, we… we have… I have, like, the script.
246 00:20:58.130 ⇒ 00:21:07.030 Casie Aviles: I need to double check, we need to double check somehow if that’s all there, because that’s, like, that’s the difficulty in, like.
247 00:21:07.030 ⇒ 00:21:08.290 Amber Lin: Yeah.
248 00:21:08.290 ⇒ 00:21:16.659 Casie Aviles: normalizing it because of these. These make it complicated, so I tried to just prompt the AI to generate the script and account for this, but…
249 00:21:17.180 ⇒ 00:21:20.700 Casie Aviles: There might be some… Next step.
250 00:21:21.080 ⇒ 00:21:22.260 Casie Aviles: are missed.
251 00:21:24.260 ⇒ 00:21:34.829 Casie Aviles: But these ones, the ones that are in tabular format, I do like spot checks, and it works pretty well for these, at least for this format, but for these, yeah, that’s…
252 00:21:35.180 ⇒ 00:21:40.399 Casie Aviles: Sometimes it’s there, sometimes it’s not, so that can blow up, like, the time of
253 00:21:41.440 ⇒ 00:21:43.170 Casie Aviles: Trying to resolve it, but…
254 00:21:43.570 ⇒ 00:21:44.840 Amber Lin: Makes sense. Okay.
255 00:21:44.920 ⇒ 00:21:50.279 Samuel Roberts: So, each of these tabs has a few of these kind of things?
256 00:21:50.610 ⇒ 00:21:51.360 Casie Aviles: Yes.
257 00:21:52.040 ⇒ 00:21:52.980 Samuel Roberts: Okay.
258 00:21:53.790 ⇒ 00:21:55.379 Samuel Roberts: Or some of them do, at least.
259 00:21:56.310 ⇒ 00:21:57.050 Casie Aviles: Yeah.
260 00:21:58.330 ⇒ 00:21:59.270 Casie Aviles: these…
261 00:22:05.460 ⇒ 00:22:06.370 Samuel Roberts: Okay.
262 00:22:11.650 ⇒ 00:22:17.160 Samuel Roberts: This data is… Crazy, okay.
263 00:22:18.460 ⇒ 00:22:19.280 Casie Aviles: Yeah.
264 00:22:24.270 ⇒ 00:22:28.700 Samuel Roberts: Alright, yeah, I’m trying to think if there’s a good way to scan that,
265 00:22:32.690 ⇒ 00:22:37.070 Samuel Roberts: Suppose there’s two things we could do here. One is to… Just…
266 00:22:37.630 ⇒ 00:22:41.089 Samuel Roberts: Look at each of the matches that are bad.
267 00:22:42.580 ⇒ 00:22:47.250 Samuel Roberts: And see if that… corresponds to…
268 00:22:47.530 ⇒ 00:22:51.420 Samuel Roberts: those ones. Actually, I imagine it will for several of them, right?
269 00:22:51.700 ⇒ 00:22:54.430 Samuel Roberts: Yeah. Okay.
270 00:22:58.790 ⇒ 00:23:02.459 Samuel Roberts: But we might still be missing ones from those tabs.
271 00:23:05.050 ⇒ 00:23:13.069 Samuel Roberts: Is that the thought? Like, we could know if we have ones, but we might be missing some. Is that the idea here? Without just checking each one of these?
272 00:23:14.040 ⇒ 00:23:18.290 Casie Aviles: Yeah, yeah, so… For example, yeah, this could be…
273 00:23:19.080 ⇒ 00:23:21.659 Casie Aviles: I can check right now, if that…
274 00:23:21.660 ⇒ 00:23:26.430 Samuel Roberts: Yeah, I mean, I think I’m just trying to think, like, we could do each one of those, I’m just trying to think if there’s a better way to do that.
275 00:23:27.180 ⇒ 00:23:29.040 Samuel Roberts: Yeah, okay.
276 00:23:38.680 ⇒ 00:23:45.060 Casie Aviles: Yeah, for example, this one wasn’t, yeah, like, it’s not… There’s no mosquito missing.
277 00:23:45.210 ⇒ 00:23:47.950 Casie Aviles: This one is for… just best.
278 00:23:49.710 ⇒ 00:23:53.160 Samuel Roberts: Right, oh, mosquito misting, I remember this, because there’s also a mosquito system.
279 00:23:55.010 ⇒ 00:23:58.539 Samuel Roberts: Is that… okay, yeah, alright. So we probably need another entry.
280 00:23:59.140 ⇒ 00:24:01.279 Samuel Roberts: For some of these, that might be missing.
281 00:24:02.870 ⇒ 00:24:03.480 Casie Aviles: Yes.
282 00:24:03.480 ⇒ 00:24:04.920 Samuel Roberts: Keyed up. Okay.
283 00:24:11.840 ⇒ 00:24:16.020 Samuel Roberts: Just trying to think if there’s another… a good way to, like, search all those names,
284 00:24:20.440 ⇒ 00:24:25.800 Samuel Roberts: Yeah, I don’t have a good idea yet, let me… Think some more…
285 00:24:26.070 ⇒ 00:24:28.149 Samuel Roberts: This is also… so, hold on, so the…
286 00:24:28.900 ⇒ 00:24:33.020 Samuel Roberts: The Brian Settles we’re seeing is this one here, right?
287 00:24:34.100 ⇒ 00:24:35.380 Casie Aviles: Yes, yes, that one.
288 00:24:35.510 ⇒ 00:24:38.770 Samuel Roberts: But we’re missing this piece of information, effectively.
289 00:24:42.640 ⇒ 00:24:43.510 Samuel Roberts: Okay.
290 00:24:46.180 ⇒ 00:24:51.960 Samuel Roberts: So we probably need… another entry to confirm… okay, okay.
291 00:24:58.980 ⇒ 00:25:03.620 Samuel Roberts: I’m just trying to see… so, Daryl… so, like, is Daryl Gentry?
292 00:25:05.520 ⇒ 00:25:08.720 Samuel Roberts: He’s not in this list, so did he make it in, you think?
293 00:25:09.050 ⇒ 00:25:13.820 Samuel Roberts: Can we check that one just to see if they’re… Yeah, okay.
294 00:25:14.650 ⇒ 00:25:16.449 Casie Aviles: I know there’s, yeah, there’s Gentry here.
295 00:25:16.450 ⇒ 00:25:22.780 Samuel Roberts: There’ll be several, right? WDI… there’s the Drywood Termite, right? San Antonio. Okay, so that makes sense, but we’re missing some. Okay.
296 00:25:29.820 ⇒ 00:25:30.670 Samuel Roberts: Okay.
297 00:25:32.570 ⇒ 00:25:36.869 Samuel Roberts: Yeah, I bet if we just look for the ones that are missing the zip counts, we’d get pretty far.
298 00:25:40.140 ⇒ 00:25:44.930 Samuel Roberts: And then we could just search names and… Confirm through the sheets, maybe?
299 00:25:45.370 ⇒ 00:25:49.219 Samuel Roberts: For the non-tabular ones.
300 00:25:51.910 ⇒ 00:25:52.650 Casie Aviles: Okay.
301 00:25:54.520 ⇒ 00:25:58.619 Casie Aviles: So I’ll do another password for these ones, the ones that are not…
302 00:25:59.320 ⇒ 00:26:08.620 Samuel Roberts: I think… I think we have to, just to make sure they’re in there, yeah. And they’re in there the right way, like, Brian Settles, we’re missing a… we’re missing this one. Daryl Gentry looks like he was fine, but…
303 00:26:09.070 ⇒ 00:26:14.539 Samuel Roberts: It looks like we saw another dialog entry for that. Okay. Yeah, I think we… I think, I don’t know…
304 00:26:14.770 ⇒ 00:26:18.780 Samuel Roberts: If you can think of a good way to make that faster, I suppose…
305 00:26:19.520 ⇒ 00:26:23.400 Samuel Roberts: I’m just trying to see if there’s anything consistent here. Probably not, right?
306 00:26:23.910 ⇒ 00:26:26.490 Samuel Roberts: The top several rows,
307 00:26:33.150 ⇒ 00:26:38.500 Samuel Roberts: Yeah, I would say prioritize the… the… the ones missing counts.
308 00:26:39.030 ⇒ 00:26:40.519 Samuel Roberts: In the normalized sheet.
309 00:26:41.960 ⇒ 00:26:46.350 Casie Aviles: Oh, gosh. Which, and I could… do you think it’s helpful if I also, like.
310 00:26:47.700 ⇒ 00:26:53.250 Casie Aviles: Right, like, generate, like, the percentage which are missing and which aren’t.
311 00:26:55.210 ⇒ 00:26:56.380 Samuel Roberts: The percentage that are…
312 00:26:57.380 ⇒ 00:27:01.160 Casie Aviles: Yeah, like, for example, how many do we have, like, that’s still missing?
313 00:27:01.930 ⇒ 00:27:02.560 Samuel Roberts: Yeah.
314 00:27:02.560 ⇒ 00:27:10.590 Amber Lin: That could be helpful, because then we can ask the clients, hey, for this person, what’s, what’s their case? Because they will know that better.
315 00:27:11.050 ⇒ 00:27:19.250 Samuel Roberts: That’s true, that’s true. If we can just get that, like, if we can, you know, run through these, confirm what we know from the sheets, and if there’s things that are still…
316 00:27:20.230 ⇒ 00:27:22.329 Samuel Roberts: Unclear, certainly, we can be like…
317 00:27:22.480 ⇒ 00:27:26.439 Samuel Roberts: Okay, this one is hard for us to understand based on the sheets, can you confirm?
318 00:27:27.060 ⇒ 00:27:28.939 Samuel Roberts: What they should be set for.
319 00:27:29.430 ⇒ 00:27:34.080 Samuel Roberts: you know, is it all of whatever, all of Austin? Is it certain zips? Is it…
320 00:27:34.260 ⇒ 00:27:36.669 Samuel Roberts: Like that, gentry guy.
321 00:27:37.060 ⇒ 00:27:42.300 Casie Aviles: It’s just, like, that one thing that we miss there. I think that that’s… yeah, I would say try to get these all…
322 00:27:42.440 ⇒ 00:27:44.380 Samuel Roberts: To zero based on what we know.
323 00:27:44.720 ⇒ 00:27:51.579 Samuel Roberts: And if we can’t do that, add that to another list kind of thing. So yeah, I would say the, like, percentages would be a good one to know.
324 00:27:53.240 ⇒ 00:27:54.370 Casie Aviles: Okay, okay.
325 00:27:55.360 ⇒ 00:27:58.700 Samuel Roberts: Especially if we’re off by a few for some people, we can then confirm with them.
326 00:28:01.420 ⇒ 00:28:07.159 Amber Lin: So, I have one question on service areas. So, for…
327 00:28:07.490 ⇒ 00:28:18.359 Amber Lin: Now that we’ve updated the surface area, so for someone who has, say, plumbing in San Antonio, are we only using zips that
328 00:28:18.500 ⇒ 00:28:23.560 Amber Lin: are serviced in San Antonio, or are we just using all zips in San Antonio?
329 00:28:26.700 ⇒ 00:28:31.639 Casie Aviles: Oh, wait, sorry, I’m not sure if I understood that correctly.
330 00:28:31.640 ⇒ 00:28:34.510 Amber Lin: Oh, I see. So if, cause…
331 00:28:34.690 ⇒ 00:28:44.569 Amber Lin: For the assignments that we use, market areas, right? How are we arriving at what zips to include in the market area?
332 00:28:44.880 ⇒ 00:28:51.530 Casie Aviles: Oh, okay, so I… well, I’m using the… this one. I’m using this…
333 00:28:51.690 ⇒ 00:28:53.560 Casie Aviles: As, like, the source of truth.
334 00:28:53.770 ⇒ 00:28:56.190 Amber Lin: Okay, awesome.
335 00:28:56.190 ⇒ 00:28:57.240 Casie Aviles: Yes, yes.
336 00:28:57.800 ⇒ 00:29:06.070 Casie Aviles: there’s, like, discrepancies even here, like, I just, like, for, like, a few minutes ago, I… there’s… I got, like, a feedback.
337 00:29:06.320 ⇒ 00:29:10.500 Casie Aviles: But yeah, I’m using this.
338 00:29:11.740 ⇒ 00:29:14.889 Casie Aviles: As, as the source of truth, but…
339 00:29:15.200 ⇒ 00:29:15.890 Amber Lin: Okay.
340 00:29:16.300 ⇒ 00:29:16.960 Casie Aviles: Good piece.
341 00:29:16.960 ⇒ 00:29:26.210 Amber Lin: So, when we use those zips, if, say, plumbing is not serviced in that zip, are we still including that in the market areas?
342 00:29:29.880 ⇒ 00:29:36.439 Casie Aviles: I… no, I think not, since I feel like… I believe, right, that this one is,
343 00:29:37.250 ⇒ 00:29:39.500 Casie Aviles: Kind of like a separate table.
344 00:29:42.230 ⇒ 00:29:46.790 Casie Aviles: Wait, sorry, I’m not sure if I… if I… Explain that.
345 00:29:46.790 ⇒ 00:29:47.109 Amber Lin: Oh, yeah.
346 00:29:47.110 ⇒ 00:29:47.910 Casie Aviles: well.
347 00:29:48.180 ⇒ 00:29:57.720 Amber Lin: So, for example, for Austin, there’s… there’s how many zips in that service… in that service area zip code?
348 00:29:57.850 ⇒ 00:30:12.910 Amber Lin: Right, but then some… for some services, it says no. So, for example, BedBug has half of it that says no. So, are we including everything in column A under Austin, or are we only including whatever BedBug says yes?
349 00:30:13.370 ⇒ 00:30:15.109 Amber Lin: In Austin.
350 00:30:16.110 ⇒ 00:30:18.830 Casie Aviles: Yeah, right, no, I’m not…
351 00:30:18.830 ⇒ 00:30:26.109 Amber Lin: We give something that says no bed bug, and we say, hey, this person serviced this zip code, then that would be a wrong answer.
352 00:30:28.190 ⇒ 00:30:29.179 Casie Aviles: Let me see…
353 00:30:29.970 ⇒ 00:30:31.940 Samuel Roberts: Right, right. So I think…
354 00:30:33.150 ⇒ 00:30:41.180 Samuel Roberts: So, can we maybe find an example of, like, someone who does bedbugs in Austin? Because then they should only really be on a few zip codes, right?
355 00:30:41.750 ⇒ 00:30:45.510 Amber Lin: I think PEST is alright, because past the assignment should be… is it.
356 00:30:45.510 ⇒ 00:30:46.090 Samuel Roberts: Okay, yeah.
357 00:30:46.090 ⇒ 00:30:47.550 Amber Lin: For my.
358 00:30:47.550 ⇒ 00:30:49.540 Samuel Roberts: Other other ones that aren’t… yeah, okay.
359 00:30:49.540 ⇒ 00:30:49.980 Casie Aviles: Clinical.
360 00:30:49.980 ⇒ 00:30:51.500 Amber Lin: Mechanical would be easier.
361 00:30:52.070 ⇒ 00:30:53.430 Samuel Roberts: Alright, so let’s…
362 00:30:58.040 ⇒ 00:30:58.540 Amber Lin: Yeah, what…
363 00:30:58.540 ⇒ 00:30:59.409 Samuel Roberts: These are…
364 00:31:00.370 ⇒ 00:31:07.630 Casie Aviles: what I’ve been doing is I’ve just been copying this and assigning the technician to these.
365 00:31:08.690 ⇒ 00:31:09.560 Amber Lin: I see.
366 00:31:22.740 ⇒ 00:31:23.550 Samuel Roberts: Okay.
367 00:31:27.810 ⇒ 00:31:28.430 Casie Aviles: Yeah.
368 00:31:28.430 ⇒ 00:31:31.849 Samuel Roberts: mechanical, there should be no appliance. What does it say there?
369 00:31:33.260 ⇒ 00:31:34.710 Samuel Roberts: Oh no, it looked at the wrong one.
370 00:31:36.490 ⇒ 00:31:39.039 Casie Aviles: Hvac. HVAC.
371 00:31:41.640 ⇒ 00:31:43.479 Casie Aviles: Let me say yes here.
372 00:31:44.340 ⇒ 00:31:50.999 Amber Lin: Yeah, maybe look at two rows above it, where it has… where it has some yeses and nos.
373 00:31:51.560 ⇒ 00:31:56.020 Amber Lin: So, 76524. Oh, yeah, that one, too.
374 00:31:57.720 ⇒ 00:31:58.730 Casie Aviles: We’ll probably do then.
375 00:31:58.730 ⇒ 00:32:04.170 Amber Lin: See, is that… is that zip code included in… in their service areas?
376 00:32:07.890 ⇒ 00:32:14.329 Amber Lin: Okay, that’s… That’s interesting. So they just don’t include it at all?
377 00:32:15.310 ⇒ 00:32:19.700 Amber Lin: But then they say, sir, they do serve as HVAC and plumbing there.
378 00:32:20.390 ⇒ 00:32:21.460 Amber Lin: So…
379 00:32:22.210 ⇒ 00:32:26.039 Casie Aviles: Yeah, that’s the… The challenge there.
380 00:32:26.040 ⇒ 00:32:33.560 Samuel Roberts: What the hell? Okay, so… They service HVAC, so is there anyone that would get this assignment right here, right now? Like, is there any way that that could have happened?
381 00:32:34.210 ⇒ 00:32:38.309 Samuel Roberts: Like, how do… how do… if they saw this, how would they know who does that?
382 00:32:38.940 ⇒ 00:32:39.740 Samuel Roberts: Without the data.
383 00:32:39.740 ⇒ 00:32:56.619 Amber Lin: I guess… I guess they don’t look at the service areas. They just know that, okay, this person is in Bell County. I’m gonna go to the service areas sheet to see if we service there. If yes, I’m gonna find this technician for Bell County, and then assign it there.
384 00:32:56.620 ⇒ 00:32:57.830 Samuel Roberts: Yeah. Because you can’t…
385 00:32:57.830 ⇒ 00:33:11.420 Amber Lin: So, we can send this as a message to confirm. Like, they have so much manual stuff, and just tons of things get shipped off in between, and then it doesn’t match up.
386 00:33:15.390 ⇒ 00:33:16.280 Amber Lin: Okay.
387 00:33:18.240 ⇒ 00:33:21.059 Amber Lin: Okay, we can, we can ask, we can ask about…
388 00:33:21.060 ⇒ 00:33:23.830 Samuel Roberts: Yeah, I think… I think, yeah, if they can clarify that, that’d be good.
389 00:33:24.160 ⇒ 00:33:24.850 Amber Lin: Okay.
390 00:33:25.290 ⇒ 00:33:26.410 Amber Lin: Yeah.
391 00:33:27.140 ⇒ 00:33:32.290 Casie Aviles: That’s also, like, another thing where… Here, it’s labeled as Austin.
392 00:33:32.710 ⇒ 00:33:37.760 Casie Aviles: So it’s an area name, but here we’re getting, like, Belgar.
393 00:33:37.760 ⇒ 00:33:38.250 Amber Lin: What?
394 00:33:38.250 ⇒ 00:33:39.369 Casie Aviles: That’s the branch.
395 00:33:39.470 ⇒ 00:33:39.970 Casie Aviles: So that’s.
396 00:33:39.970 ⇒ 00:33:42.289 Amber Lin: Oh my god, no!
397 00:33:42.290 ⇒ 00:33:44.269 Casie Aviles: Makes it more complicated, like…
398 00:33:44.270 ⇒ 00:33:58.279 Amber Lin: Okay, I think I’m gonna ask them if we can just use the service area sheet as our source of truth, because that’s the tabular format that we want. So I’m just gonna confirm, hey, is this…
399 00:33:58.400 ⇒ 00:34:04.530 Amber Lin: Still true? If so, we’re gonna ignore your mechanical service areas and use this.
400 00:34:05.910 ⇒ 00:34:17.219 Amber Lin: Okay, I’ll ask them. Okay, let’s, you can ignore this for now, we’ll do the regular validation, as Sam said, and if they get back to us, this will be another ticket.
401 00:34:17.909 ⇒ 00:34:18.569 Samuel Roberts: Yeah.
402 00:34:19.070 ⇒ 00:34:19.500 Amber Lin: Okay.
403 00:34:19.500 ⇒ 00:34:20.010 Casie Aviles: Okay.
404 00:34:20.310 ⇒ 00:34:26.319 Amber Lin: Sounds good. That’s all the zip code stuff, right? You’re… you’re just missing the inspector sheet, or is Apple Showtime?
405 00:34:27.560 ⇒ 00:34:33.079 Casie Aviles: The inspector’s already here, you just need to do, like, another pass with the…
406 00:34:33.080 ⇒ 00:34:33.770 Amber Lin: Okay.
407 00:34:33.770 ⇒ 00:34:37.130 Casie Aviles: With the zip count matches that are… or mismatches.
408 00:34:37.550 ⇒ 00:34:42.779 Amber Lin: I see, sounds good. Are we planning to use any of the triage tickets in our testing?
409 00:34:43.560 ⇒ 00:34:50.249 Casie Aviles: Yeah, that’s the plan. That’s also, like, what I was trying to build out, which is the automated testing.
410 00:34:50.540 ⇒ 00:34:53.540 Amber Lin: Okay, okay, sounds good. I like that plan.
411 00:34:53.980 ⇒ 00:34:57.500 Amber Lin: Anything else on zip codes before we talk about the central dock?
412 00:34:58.970 ⇒ 00:35:01.449 Casie Aviles: I think that’s… that’s all I had right now.
413 00:35:01.660 ⇒ 00:35:02.270 Amber Lin: Okay.
414 00:35:02.590 ⇒ 00:35:12.849 Amber Lin: Yeah, so if I go ahead. I read your doc a little bit, but, I think it would be helpful for other folks to get context if they don’t yet, but I know you guys might have already talked.
415 00:35:16.330 ⇒ 00:35:19.410 Mustafa Raja: Yes, give me a moment, please. I’m pulling up the ticket.
416 00:35:23.440 ⇒ 00:35:24.450 Mustafa Raja: Oh…
417 00:35:42.540 ⇒ 00:35:53.410 Mustafa Raja: Okay, so, we are thinking of, dividing this… dividing the whole, central dog into this general, format, so we’ll have.
418 00:35:53.410 ⇒ 00:36:04.530 Mustafa Raja: On the top of dock header. This is just metadata on, who’s changing what, and when it’s being changed. This is quick reference for,
419 00:36:04.540 ⇒ 00:36:06.390 Mustafa Raja: CSRs, you know?
420 00:36:06.390 ⇒ 00:36:30.119 Mustafa Raja: What we treat, what we don’t treat, and stuff like that. And then this is service definitions for all services that we would offer. We would have 3 sections for all of those. Overview, covered, not covered, so what’s covered in the service and what’s not covered in the service, pricing and duration, and this would be a table with all, with all prior changes. So AI would know, okay, what’s the latest one.
421 00:36:32.140 ⇒ 00:36:41.640 Mustafa Raja: And then service code. So this, this, format is going to be forward for every service that we would, provide. For example.
422 00:36:41.890 ⇒ 00:36:53.750 Mustafa Raja: GPC, and then rodent, and then thermite bug, bed bug, and mosquito, and other sources will have… will all have these, these fields in them.
423 00:36:54.090 ⇒ 00:37:02.530 Mustafa Raja: And then this is just, their call logs or workflows that they need. And this,
424 00:37:03.240 ⇒ 00:37:25.770 Mustafa Raja: This would be, so what this document additionally says is that, we shouldn’t be, keeping, personal information in the document. That should be, taken care of by, by the zip tools, because when I communicated with it, I gave a full context on how Andy is working, so the agent says.
425 00:37:25.770 ⇒ 00:37:34.400 Mustafa Raja: That, inspectors or personnel that are in, Central Dock should be moved to, database.
426 00:37:34.860 ⇒ 00:37:38.170 Mustafa Raja: Yeah, oh…
427 00:37:38.630 ⇒ 00:37:45.180 Mustafa Raja: That’s pretty much it, and then there’s a strategy for, us, to, you know, have shared,
428 00:37:45.310 ⇒ 00:37:46.980 Mustafa Raja: I have a document where you’ll put
429 00:37:47.190 ⇒ 00:37:53.530 Mustafa Raja: All of the shared, definitions, and we could just, rather than,
430 00:37:53.630 ⇒ 00:38:08.289 Mustafa Raja: what we were suggesting before is we could use a UUID, to sort of, push, all the data, and this is… the agent suggested we should rather use, human-readable definitions, or human-readable IDs.
431 00:38:08.360 ⇒ 00:38:19.059 Mustafa Raja: So, it should look like this, and as we go beneath, we get a deeper analysis of what each section would be doing.
432 00:38:19.340 ⇒ 00:38:20.370 Amber Lin: I see.
433 00:38:20.610 ⇒ 00:38:21.410 Mustafa Raja: Okay.
434 00:38:21.410 ⇒ 00:38:24.500 Amber Lin: And I have a question here.
435 00:38:24.500 ⇒ 00:38:25.050 Mustafa Raja: Yeah, yeah, yeah.
436 00:38:25.050 ⇒ 00:38:27.359 Amber Lin: Can we go back to the sections?
437 00:38:27.550 ⇒ 00:38:32.400 Amber Lin: For the second section, you said it will be in a table.
438 00:38:32.680 ⇒ 00:38:33.159 Mustafa Raja: Let me real.
439 00:38:33.160 ⇒ 00:38:35.259 Amber Lin: I know that Andy can’t read…
440 00:38:35.500 ⇒ 00:38:43.259 Mustafa Raja: Not everything will be in a table. You see, overview, covered, not covered, these, all of these would be in… what’s it called?
441 00:38:43.390 ⇒ 00:38:44.380 Amber Lin: Huh.
442 00:38:44.800 ⇒ 00:38:56.089 Mustafa Raja: And then the pricing only would be in tables, and then, yeah, for your question… Yeah. Can Andy read tables? I would like to have it in tables, I just… Yeah.
443 00:38:56.090 ⇒ 00:38:58.600 Amber Lin: I’ve had situations where you could not read tables.
444 00:38:58.600 ⇒ 00:39:01.049 Casie Aviles: If it’s marked down, I think it can read now.
445 00:39:01.050 ⇒ 00:39:01.740 Mustafa Raja: Oh, yeah.
446 00:39:02.410 ⇒ 00:39:15.750 Mustafa Raja: If it’s marked down, it can, and then what we are doing is we are getting all of the data in JSON format from Google Docs, right? So we can convert it into whatever format we would want.
447 00:39:16.520 ⇒ 00:39:17.380 Mustafa Raja: You know?
448 00:39:19.180 ⇒ 00:39:20.069 Mustafa Raja: Does that make sense?
449 00:39:20.070 ⇒ 00:39:20.590 Samuel Roberts: Huh.
450 00:39:23.060 ⇒ 00:39:24.670 Mustafa Raja: Yeah, I think there’s also…
451 00:39:24.670 ⇒ 00:39:28.469 Samuel Roberts: I think there’s also a chance that, the new models will probably handle it better anyway.
452 00:39:28.470 ⇒ 00:39:31.130 Mustafa Raja: Oh, yeah, that’s also a good point.
453 00:39:31.130 ⇒ 00:39:33.510 Samuel Roberts: Like, we’re using a pretty outdated model still, because…
454 00:39:33.830 ⇒ 00:39:41.830 Mustafa Raja: Yeah, and this is to make sure that, you know, we are always up to date with the new pricing, and model also knows, okay.
455 00:39:42.170 ⇒ 00:39:47.619 Mustafa Raja: If anywhere else some price comes up, it would know, okay, this is something outdated.
456 00:39:49.110 ⇒ 00:39:54.279 Casie Aviles: Yeah, before, I think we couldn’t, like, export it to Markdown, right? The central.
457 00:39:54.280 ⇒ 00:40:04.300 Mustafa Raja: We are not still able to export it in Markdown. We are exporting it in JSON, and from JSON, we can convert it into whatever, right?
458 00:40:04.300 ⇒ 00:40:11.259 Casie Aviles: Oh, because… because I can… yeah, like, in the file, when you try to download it, I think there’s, like, a markdown option.
459 00:40:11.260 ⇒ 00:40:17.800 Mustafa Raja: Yeah, it’s there, but when we try to download it using the API, that’s just…
460 00:40:17.800 ⇒ 00:40:19.210 Casie Aviles: Okay.
461 00:40:22.040 ⇒ 00:40:26.770 Mustafa Raja: Yeah, so API does only provide us with JSON?
462 00:40:27.450 ⇒ 00:40:35.560 Mustafa Raja: But yeah, we can format the JSON however we want, and we are formatting it right now. I think we are already formatting it in JSON, or sorry, in Markdown already.
463 00:40:37.500 ⇒ 00:40:38.190 Samuel Roberts: I think so.
464 00:40:38.650 ⇒ 00:40:49.059 Mustafa Raja: Yeah, I believe. I believe we made it, so… I don’t think that we might have taken care of the tables, but if you’re moving with this format, I can write a script for that.
465 00:40:53.100 ⇒ 00:40:55.659 Mustafa Raja: So, yeah.
466 00:40:56.880 ⇒ 00:41:05.910 Amber Lin: Let’s see. So… What’s the gap between this and a newly generated central software there review?
467 00:41:07.470 ⇒ 00:41:08.960 Amber Lin: What do we need to do?
468 00:41:09.010 ⇒ 00:41:18.269 Mustafa Raja: Yeah, so it also generated us a table of how we are going to, you know, migrate this current pest dock. Yeah, so here…
469 00:41:18.850 ⇒ 00:41:24.240 Mustafa Raja: Here, in this compatibility section, it tells us, okay, which section goes where.
470 00:41:25.220 ⇒ 00:41:26.050 Mustafa Raja: I know?
471 00:41:27.120 ⇒ 00:41:36.350 Mustafa Raja: So services… so this heading would go under SOP workflows, same… similar for this, then pest would be going under service definitions.
472 00:41:37.130 ⇒ 00:41:42.890 Mustafa Raja: So it has a… so it also gave us a migration,
473 00:41:43.170 ⇒ 00:41:46.559 Mustafa Raja: Script also, on how we can migrate to this.
474 00:41:48.900 ⇒ 00:42:00.479 Amber Lin: Okay, that sounds good. And so, we would do… we would move things around first, and then look at if any of them have a lot of duplicate content, and then size them down, right?
475 00:42:00.950 ⇒ 00:42:01.810 Mustafa Raja: Yeah.
476 00:42:04.320 ⇒ 00:42:17.439 Mustafa Raja: Yeah, we could have one… one document that’s just, you know, what’s it called? That’s just a… that’s just… that is just shared sections among all of the,
477 00:42:17.890 ⇒ 00:42:36.589 Mustafa Raja: central docs with IDs like, with IDs like this, and then, we just, you know, replace the duplicates with these IDs, and just as we are about to embed these sections, these IDs would get replaced by the actual content.
478 00:42:38.060 ⇒ 00:42:40.830 Amber Lin: Okay, where do you guys think?
479 00:42:41.550 ⇒ 00:42:51.140 Amber Lin: Do you think we’re good to move forward to make a test section and test on a dock, or any cautions or risks you guys see?
480 00:42:52.740 ⇒ 00:42:58.399 Samuel Roberts: No, I think… I think it’s good to go. I kind of wanted your eyes on it first, before I was like, yeah, it’s ready, but.
481 00:42:58.990 ⇒ 00:42:59.350 Samuel Roberts: I think.
482 00:42:59.350 ⇒ 00:43:00.020 Amber Lin: Okay.
483 00:43:00.040 ⇒ 00:43:01.209 Samuel Roberts: I think it’s got everything…
484 00:43:01.210 ⇒ 00:43:03.360 Amber Lin: I’d love to have something to show them.
485 00:43:03.720 ⇒ 00:43:10.450 Samuel Roberts: Yeah, I think if we can… we can run that on this. The other thing I think we talked about, Mustafa, was ragging it slightly differently by sections instead of by…
486 00:43:10.450 ⇒ 00:43:11.770 Mustafa Raja: Oh, yeah, yeah.
487 00:43:11.770 ⇒ 00:43:13.540 Samuel Roberts: Which I think will also help a lot.
488 00:43:13.690 ⇒ 00:43:15.459 Mustafa Raja: Yeah. Once it’s cleaned up.
489 00:43:16.330 ⇒ 00:43:17.070 Mustafa Raja: For context.
490 00:43:17.070 ⇒ 00:43:18.380 Samuel Roberts: Do you think the doc should still…
491 00:43:18.380 ⇒ 00:43:35.660 Mustafa Raja: Yeah, for context, what’s happening right now is, AI has, sorry, the embedding pipeline has a limit of characters it can, embed in a moment. So what happens is, if a section is bigger than that limit, it gets split into two sections.
492 00:43:35.660 ⇒ 00:43:44.410 Mustafa Raja: Right, so a single section would become two sections, and… or two or more sections, and then that might just create some duplicates within embeddings also.
493 00:43:45.280 ⇒ 00:43:57.459 Mustafa Raja: Since we have, an overlap between the sections too. So, I had an idea, that we should rather, move from NITM for embedding.
494 00:43:57.530 ⇒ 00:44:06.170 Mustafa Raja: move to Mastra, and then Mastra hopefully does provide us, capability to embed the whole section
495 00:44:06.780 ⇒ 00:44:10.959 Mustafa Raja: Altogether. So hopefully, that might improve our results too.
496 00:44:13.450 ⇒ 00:44:15.970 Samuel Roberts: Yeah, yeah, I think once this is cleaned up, that’ll be great.
497 00:44:16.170 ⇒ 00:44:17.979 Mustafa Raja: Yeah, yeah, I agree.
498 00:44:20.570 ⇒ 00:44:23.250 Mustafa Raja: So, what does the team think about this?
499 00:44:23.700 ⇒ 00:44:25.049 Samuel Roberts: Yeah, I say go for it.
500 00:44:25.300 ⇒ 00:44:25.970 Mustafa Raja: Okay.
501 00:44:26.450 ⇒ 00:44:28.180 Amber Lin: How long do you think it’ll take?
502 00:44:29.870 ⇒ 00:44:43.199 Mustafa Raja: We already have, what’s it called? Embedding thing, right, in the… in NHN, so I’ll translate that, I’ll make sure that we are able to cater to, these… what’s it called? These tables.
503 00:44:43.950 ⇒ 00:44:46.020 Samuel Roberts: Do you mean the new… generating the new dock?
504 00:44:47.050 ⇒ 00:44:47.830 Mustafa Raja: Sorry?
505 00:44:48.210 ⇒ 00:44:50.110 Samuel Roberts: I think, Amber, do you mean generating the new doctor?
506 00:44:50.110 ⇒ 00:44:55.019 Mustafa Raja: Yeah, for generating the new doc, I’ll just follow the strategy that it’s giving us.
507 00:44:57.770 ⇒ 00:45:00.420 Samuel Roberts: Okay. Yeah, I was kind of thinking about trying to see.
508 00:45:00.420 ⇒ 00:45:13.649 Amber Lin: I assume this new dock would… would not have… let’s… let’s see… it wouldn’t be any shorter than the current central doc, right? So it will just copy word for word and just move them to the right section.
509 00:45:13.860 ⇒ 00:45:15.300 Mustafa Raja: Yeah, that’s pretty much it.
510 00:45:15.300 ⇒ 00:45:21.249 Amber Lin: Okay, yeah, that’s… that’s good. And then we can do the size-down duplicate stuff later.
511 00:45:21.710 ⇒ 00:45:22.290 Samuel Roberts: Yeah.
512 00:45:23.140 ⇒ 00:45:24.050 Mustafa Raja: technical data.
513 00:45:24.440 ⇒ 00:45:31.029 Amber Lin: Sounds good. Can we have an initial review by tomorrow? Not by start of day, but just sometime.
514 00:45:31.030 ⇒ 00:45:35.439 Mustafa Raja: Yeah, tomorrow, end of day, end of day would be nice, yeah. Tomorrow, end of day would be nice.
515 00:45:35.440 ⇒ 00:45:38.820 Amber Lin: Can I have it before my meeting with the clients?
516 00:45:38.820 ⇒ 00:45:40.029 Mustafa Raja: When’s your meeting?
517 00:45:40.030 ⇒ 00:45:46.750 Amber Lin: So that would be around… Like, an hour later from now, so it’s 12 PM PST.
518 00:45:46.860 ⇒ 00:45:48.899 Amber Lin: So, like, bye!
519 00:45:49.130 ⇒ 00:45:51.790 Amber Lin: An hour later, tomorrow.
520 00:45:52.490 ⇒ 00:45:53.440 Mustafa Raja: Yeah, exactly.
521 00:45:53.440 ⇒ 00:45:54.040 Amber Lin: people.
522 00:45:54.300 ⇒ 00:45:59.040 Amber Lin: It doesn’t have to be polished, I just want to show them, like, hey, we are doing something, and then
523 00:46:00.050 ⇒ 00:46:03.520 Amber Lin: where I need to show them the exact things and go through the text.
524 00:46:03.920 ⇒ 00:46:05.399 Mustafa Raja: Okay, I’ll prioritize this.
525 00:46:06.210 ⇒ 00:46:07.480 Mustafa Raja: For ABT, then?
526 00:46:08.200 ⇒ 00:46:08.770 Amber Lin: Okay.
527 00:46:10.000 ⇒ 00:46:10.820 Amber Lin: Cool.
528 00:46:12.120 ⇒ 00:46:18.020 Mustafa Raja: So, which drive does the new dog calls in? Is it our own drive, or ABC drive?
529 00:46:18.440 ⇒ 00:46:24.229 Amber Lin: I mean, we can do it in our own drive for now, because there’s… Yeah, definitely.
530 00:46:24.690 ⇒ 00:46:27.970 Mustafa Raja: Okay, okay, yeah, I’ll do that in Brainforce Drive then.
531 00:46:30.790 ⇒ 00:46:31.910 Amber Lin: Awesome.
532 00:46:32.380 ⇒ 00:46:35.789 Amber Lin: Anything else on ABC?
533 00:46:36.540 ⇒ 00:46:41.599 Amber Lin: Have we… oh, I guess, have we had time to look at triage tickets at all?
534 00:46:43.340 ⇒ 00:46:48.389 Mustafa Raja: I haven’t… I haven’t been able to find time. I’ve been packed with reading stuff.
535 00:46:49.030 ⇒ 00:46:54.090 Amber Lin: Understandable. I, I think all of, all of you are packed, so… Thanks.
536 00:46:54.240 ⇒ 00:46:59.139 Amber Lin: I guess if we have… If we have a few moments…
537 00:46:59.360 ⇒ 00:47:09.350 Amber Lin: Can we talk about strategies of how can we reduce time that we spend on triage tickets? Like, what do we do that takes time on triage tickets? Is it…
538 00:47:09.520 ⇒ 00:47:11.729 Amber Lin: assigning them? Is it…
539 00:47:11.890 ⇒ 00:47:19.810 Amber Lin: Finding where they are in the docs, like, what would make it take less time for us?
540 00:47:20.100 ⇒ 00:47:21.020 Mustafa Raja: Hmm…
541 00:47:21.020 ⇒ 00:47:23.370 Casie Aviles: I think part of, part of the…
542 00:47:23.490 ⇒ 00:47:30.290 Casie Aviles: what could speed it up was also, like, the rationale behind, you know, the automated testing that I’m thinking about.
543 00:47:30.290 ⇒ 00:47:30.880 Amber Lin: Hmm.
544 00:47:31.970 ⇒ 00:47:40.200 Casie Aviles: That’s… that also keeps the triage tickets in mind, because the idea is we want to collect, like, the… the questions there, and then…
545 00:47:40.980 ⇒ 00:47:48.060 Casie Aviles: We go through, like… Or we re-enter, basically, the questions to Andy, and then…
546 00:47:48.510 ⇒ 00:47:50.580 Casie Aviles: See, like, if the output has improved.
547 00:47:51.440 ⇒ 00:47:59.120 Amber Lin: I see, okay, so that will help reduce, especially the older, stale triage tickets, so we can take out quite a few in one pass.
548 00:48:00.010 ⇒ 00:48:02.340 Casie Aviles: Yeah, yeah, that’s, like, the idea behind it.
549 00:48:02.990 ⇒ 00:48:03.880 Amber Lin: Okay.
550 00:48:04.330 ⇒ 00:48:13.220 Amber Lin: What about, say, new triage tickets that… what are the steps we need to take when we… when we work on them? What do you guys do?
551 00:48:15.560 ⇒ 00:48:18.459 Casie Aviles: When there are new triage tickets,
552 00:48:18.880 ⇒ 00:48:26.000 Casie Aviles: Usually, I… well, I see them first in the triage section, and then it gets assigned, I think, by
553 00:48:26.810 ⇒ 00:48:27.750 Casie Aviles: Jenny’s…
554 00:48:29.330 ⇒ 00:48:34.640 Casie Aviles: to me, for example, it gets assigned to me, and then when she assigns it to me, that’s when…
555 00:48:34.970 ⇒ 00:48:37.299 Casie Aviles: I have to look through each…
556 00:48:37.660 ⇒ 00:48:39.859 Casie Aviles: Take it, like, one… one by one.
557 00:48:40.430 ⇒ 00:48:43.780 Casie Aviles: Where I have to…
558 00:48:43.900 ⇒ 00:48:51.680 Casie Aviles: you know, I have to get, like, I have to recreate, basically, the issue that they were… Experiencing.
559 00:48:52.560 ⇒ 00:48:59.420 Casie Aviles: And I guess manually testing that one by one with Andy is what takes the time, so that’s…
560 00:48:59.920 ⇒ 00:49:05.960 Amber Lin: Okay, so that will also be helped by our new, like, the automated testing thing.
561 00:49:05.960 ⇒ 00:49:10.660 Casie Aviles: Yeah, yeah, that’s also, like, part of it. I mean, there are other things that…
562 00:49:11.130 ⇒ 00:49:28.980 Casie Aviles: I do that as well, but I’m not sure, like, if we can automate that yet, like, for example, looking for, like, the root issue, I’m not sure yet, but at least, like, the manual input, text input to Andy, send it, and then wait for response. That should not… that should be resolved with the automated test.
563 00:49:28.980 ⇒ 00:49:37.839 Amber Lin: Okay. So, do you plan to have that in a UI, or… I feel like it might not be the best to have it in worksheets.
564 00:49:38.010 ⇒ 00:49:39.069 Amber Lin: If that’s…
565 00:49:41.000 ⇒ 00:49:42.380 Casie Aviles: Oh,
566 00:49:42.900 ⇒ 00:49:50.199 Casie Aviles: Right now, like, the initial plan, at least for the first version, is to just have it, like, as a workflow.
567 00:49:51.400 ⇒ 00:49:57.269 Casie Aviles: So we’re going to, code it via cursor, and
568 00:49:57.600 ⇒ 00:50:00.980 Casie Aviles: Yeah, like, the initial version is just on Google Sheets, but…
569 00:50:01.690 ⇒ 00:50:12.580 Casie Aviles: Maybe we… yeah, maybe in the future we could do, like, another, like, dedicated UI for it, if you need to, so it can be more usable.
570 00:50:14.370 ⇒ 00:50:16.760 Casie Aviles: Alright, yeah, that’s… that’s just…
571 00:50:17.300 ⇒ 00:50:21.979 Casie Aviles: I don’t have, like, a plan yet for that kind of, like, formalized UI, but…
572 00:50:22.470 ⇒ 00:50:22.980 Amber Lin: Okay.
573 00:50:23.400 ⇒ 00:50:30.209 Amber Lin: Do you have, like, an easy way to find out the logs of, say, a question? Like, what…
574 00:50:30.530 ⇒ 00:50:36.500 Amber Lin: Do you have to… is it hard to navigate to what happens in the back end?
575 00:50:40.200 ⇒ 00:50:42.100 Casie Aviles: You mean, like, with Andy?
576 00:50:42.150 ⇒ 00:50:49.459 Amber Lin: Yeah, like, for example, triage comes in, and you said you have to, one, look it up, look the question up, and then see how it executed.
577 00:50:49.500 ⇒ 00:50:52.069 Casie Aviles: Yes, yes. Is that process time-consuming?
578 00:50:52.810 ⇒ 00:50:59.600 Casie Aviles: Yeah, that… I have to go to NATEN, look at the execution log, and then check which node
579 00:51:00.050 ⇒ 00:51:09.310 Casie Aviles: Basically, which node failed, or… yeah, if it’s the db query tool, or is it somewhere in the prompt? That’s another manual step that I do.
580 00:51:12.130 ⇒ 00:51:14.539 Amber Lin: Team, what do you think? Do you think there’s.
581 00:51:14.540 ⇒ 00:51:15.380 Samuel Roberts: Yeah, I’m wondering…
582 00:51:15.380 ⇒ 00:51:16.830 Amber Lin: Can I make that easier?
583 00:51:17.920 ⇒ 00:51:22.080 Samuel Roberts: Are we… we’re not sharing which execution it was in the triage?
584 00:51:22.200 ⇒ 00:51:25.810 Samuel Roberts: Or you have to… it’s just the… actually getting to the execution date on that end.
585 00:51:26.050 ⇒ 00:51:26.930 Samuel Roberts: Navin.
586 00:51:28.060 ⇒ 00:51:29.290 Samuel Roberts: D.
587 00:51:29.520 ⇒ 00:51:30.360 Samuel Roberts: Damn.
588 00:51:30.850 ⇒ 00:51:33.830 Casie Aviles: Sorry? I think, I think you’re… is it just me, or…
589 00:51:33.830 ⇒ 00:51:35.340 Amber Lin: was cut.
590 00:51:35.340 ⇒ 00:51:36.690 Samuel Roberts: out a little bit.
591 00:51:36.920 ⇒ 00:51:38.139 Samuel Roberts: Computer froze.
592 00:51:38.610 ⇒ 00:51:40.170 Amber Lin: Oh, I can hear you now.
593 00:51:41.070 ⇒ 00:51:46.190 Samuel Roberts: Yeah, okay, yeah, sorry, my computer just froze for a second, I don’t know what happened there.
594 00:51:46.340 ⇒ 00:51:48.070 Samuel Roberts: I was saying, is it the…
595 00:51:48.340 ⇒ 00:51:52.060 Samuel Roberts: Are we sending the execution from N8N into the triage ticket?
596 00:51:53.470 ⇒ 00:51:56.040 Casie Aviles: No, no, no, it’s not yet being sent.
597 00:51:56.620 ⇒ 00:51:59.960 Samuel Roberts: Okay, is that something we… cause that we…
598 00:52:00.690 ⇒ 00:52:01.210 Mustafa Raja: Right.
599 00:52:01.210 ⇒ 00:52:03.110 Samuel Roberts: Can we add something in Snowflake for that?
600 00:52:03.110 ⇒ 00:52:09.620 Mustafa Raja: Yeah, it’s being smashed to Snowflake, I believe, but not… Okay, so you should… so would that… would it help to update what…
601 00:52:09.700 ⇒ 00:52:12.410 Samuel Roberts: from Snowflake to the triage in linear.
602 00:52:13.260 ⇒ 00:52:13.759 Mustafa Raja: We could be.
603 00:52:13.760 ⇒ 00:52:14.399 Samuel Roberts: That makes…
604 00:52:14.400 ⇒ 00:52:14.980 Mustafa Raja: relief.
605 00:52:15.550 ⇒ 00:52:19.369 Samuel Roberts: Yeah, would that make it at least a step faster to find that?
606 00:52:19.650 ⇒ 00:52:21.119 Samuel Roberts: Broken note or whatever.
607 00:52:25.550 ⇒ 00:52:27.200 Casie Aviles: Yeah, I mean.
608 00:52:27.200 ⇒ 00:52:27.970 Samuel Roberts: Or is that…
609 00:52:28.400 ⇒ 00:52:33.720 Casie Aviles: Well, it’s just gonna help with, like, looking for the exact execution log.
610 00:52:34.250 ⇒ 00:52:41.169 Samuel Roberts: Yeah, I think we should add that then, that’s definitely worth doing. Now, I think we added that in Snowflake at some point, but I don’t think we added it to the…
611 00:52:41.170 ⇒ 00:52:41.649 Casie Aviles: Yeah, that…
612 00:52:41.650 ⇒ 00:52:42.650 Samuel Roberts: message.
613 00:52:42.790 ⇒ 00:52:43.360 Casie Aviles: Okay.
614 00:52:43.600 ⇒ 00:52:51.230 Amber Lin: I remember Utam talking about this when we had major, like, zip code issues in December. Is this something we’re happy to advance from?
615 00:52:54.040 ⇒ 00:52:55.109 Samuel Roberts: Say that again?
616 00:52:55.320 ⇒ 00:52:59.019 Amber Lin: Like, will we have this in menstruel, or is this just an AN issue?
617 00:52:59.210 ⇒ 00:53:03.619 Samuel Roberts: This isn’t an issue with the way it does the, execution logs.
618 00:53:03.950 ⇒ 00:53:06.439 Samuel Roberts: With Maestro, we’re logging that to…
619 00:53:06.650 ⇒ 00:53:11.450 Samuel Roberts: the SQLite database, which is then getting logged to the cloud SQL, isn’t that right, Mustava?
620 00:53:11.450 ⇒ 00:53:16.260 Mustafa Raja: Oh, yeah, we just removed the SQLite, you know?
621 00:53:16.260 ⇒ 00:53:20.699 Samuel Roberts: So, yeah, we were, yeah. The execution log should be all there, so…
622 00:53:20.700 ⇒ 00:53:31.069 Amber Lin: I see. So, if… I mean, if the migration happens in a week, I think, and if we’ll be there after migration, I think we can save us some time and not do it.
623 00:53:31.350 ⇒ 00:53:36.970 Amber Lin: For now, but… Like, if… maybe we have to backfill?
624 00:53:37.950 ⇒ 00:53:38.849 Amber Lin: Triage tickets?
625 00:53:38.850 ⇒ 00:53:39.230 Samuel Roberts: Maybe.
626 00:53:39.230 ⇒ 00:53:39.870 Amber Lin: those.
627 00:53:39.870 ⇒ 00:53:44.889 Samuel Roberts: I don’t know, I don’t think it would be that heavy of a lift to add that from the Snowflake to…
628 00:53:45.070 ⇒ 00:53:46.080 Amber Lin: Hmm, okay.
629 00:53:46.660 ⇒ 00:53:49.530 Samuel Roberts: the linear, if that would be advantageous for you, Casey.
630 00:53:51.220 ⇒ 00:53:51.950 Casie Aviles: Yeah.
631 00:53:51.950 ⇒ 00:53:52.980 Samuel Roberts: But you tell me.
632 00:53:55.980 ⇒ 00:54:04.759 Casie Aviles: I think we could… we could just add, like, the execution, at least, you know, to the tickets and the messages in Slack, so…
633 00:54:05.160 ⇒ 00:54:07.520 Casie Aviles: Think we can do that, please?
634 00:54:08.280 ⇒ 00:54:09.320 Samuel Roberts: Okay, that sounds good.
635 00:54:11.380 ⇒ 00:54:12.150 Amber Lin: Alright.
636 00:54:12.380 ⇒ 00:54:14.519 Casie Aviles: I think that should be faster.
637 00:54:15.140 ⇒ 00:54:15.520 Samuel Roberts: Yeah.
638 00:54:15.520 ⇒ 00:54:16.530 Casie Aviles: Small lift.
639 00:54:17.320 ⇒ 00:54:17.900 Samuel Roberts: Okay.
640 00:54:20.650 ⇒ 00:54:25.720 Samuel Roberts: Yeah, if we can add a ticket for that, I can even take a look at it if you need, while you’re doing the, zip code stuff.
641 00:54:26.610 ⇒ 00:54:37.299 Amber Lin: Cool. So, zip… so, I think we’re at the end of the meeting. For zip codes, I think Casey’s doing the QA, Central Block and stuff I was doing…
642 00:54:37.850 ⇒ 00:54:41.260 Amber Lin: the… the doc,
643 00:54:41.600 ⇒ 00:54:49.340 Amber Lin: run through. I will ask about market areas, so I’ll make a ticket for the execution logs for later.
644 00:54:49.680 ⇒ 00:55:00.689 Amber Lin: And then… I think for the migration, We need… Staging, something about the models…
645 00:55:01.700 ⇒ 00:55:07.859 Amber Lin: And we need, the… Triage and feedback, thumbs up, thumbs up and stuff.
646 00:55:09.220 ⇒ 00:55:13.210 Samuel Roberts: For the models, I think we’re sending that message to Tim, right, Mustafa? Once you have it?
647 00:55:13.690 ⇒ 00:55:14.340 Samuel Roberts: Hmm.
648 00:55:14.340 ⇒ 00:55:22.129 Mustafa Raja: Yeah, I will be, you know, revisiting it, just to make sure that, you know, there’s no vertex yet in there.
649 00:55:22.660 ⇒ 00:55:23.959 Samuel Roberts: Okay, cool, sounds good.
650 00:55:23.960 ⇒ 00:55:25.490 Amber Lin: Sounds good, yeah, less solar.
651 00:55:25.490 ⇒ 00:55:25.810 Samuel Roberts: Nice.
652 00:55:25.960 ⇒ 00:55:27.929 Amber Lin: Fantastic. Alright, thanks everyone.
653 00:55:27.930 ⇒ 00:55:28.969 Mustafa Raja: Thank you, have a good day.
654 00:55:28.970 ⇒ 00:55:29.570 Samuel Roberts: Alright, yeah.
655 00:55:29.570 ⇒ 00:55:30.020 Casie Aviles: Thank you.
656 00:55:30.020 ⇒ 00:55:30.580 Samuel Roberts: one.