Meeting Title: ABC Standup Date: 2025-10-29 Meeting participants: Casie Aviles, Samuel Roberts, Mustafa Raja, Amber Lin
WEBVTT
1 00:00:15.320 ⇒ 00:00:16.030 Mustafa Raja: Hey…
2 00:00:17.340 ⇒ 00:00:21.849 Samuel Roberts: Hello, hello, alright. We’re waiting on Amber, right?
3 00:00:22.980 ⇒ 00:00:23.650 Mustafa Raja: Yeah.
4 00:00:24.600 ⇒ 00:00:25.010 Samuel Roberts: Correct.
5 00:00:25.010 ⇒ 00:00:28.960 Mustafa Raja: I guess, Regarding the spike.
6 00:00:29.200 ⇒ 00:00:38.450 Mustafa Raja: We already have the central doc in one… for every department, the embeddings get stored.
7 00:00:38.560 ⇒ 00:00:45.740 Mustafa Raja: Put in one single table that then has metadata fields based on those departments.
8 00:00:46.640 ⇒ 00:00:53.330 Mustafa Raja: Right. Like, we can go through, or let me know if we should, refund.
9 00:00:53.330 ⇒ 00:00:56.359 Samuel Roberts: Yeah, let’s hold… let’s wait just a minute and make sure…
10 00:00:56.800 ⇒ 00:00:57.550 Mustafa Raja: Okay, okay.
11 00:00:57.710 ⇒ 00:00:58.659 Mustafa Raja: Now, that’s true.
12 00:01:13.570 ⇒ 00:01:16.183 Samuel Roberts: Got it.
13 00:01:29.890 ⇒ 00:01:32.470 Samuel Roberts: Okay, I’ll need to message her real quick.
14 00:02:04.330 ⇒ 00:02:07.320 Samuel Roberts: Okay, let me… I don’t responded here.
15 00:02:11.940 ⇒ 00:02:13.239 Samuel Roberts: Why is my son?
16 00:02:18.940 ⇒ 00:02:21.329 Samuel Roberts: Cool, okay, good. I’ll type the message.
17 00:02:23.430 ⇒ 00:02:24.430 Amber Lin: Hi!
18 00:02:24.910 ⇒ 00:02:25.920 Samuel Roberts: Hello.
19 00:02:26.330 ⇒ 00:02:29.500 Amber Lin: Okay, we have a few stuff.
20 00:02:29.750 ⇒ 00:02:35.630 Amber Lin: Let’s get started… Okay.
21 00:02:46.960 ⇒ 00:02:52.570 Amber Lin: So first, there is… Alright, sorry.
22 00:02:54.140 ⇒ 00:02:59.490 Amber Lin: There’s a long cruise, Reminders and loading.
23 00:03:00.150 ⇒ 00:03:01.660 Amber Lin: indicators.
24 00:03:03.690 ⇒ 00:03:04.380 Casie Aviles: Yes.
25 00:03:08.310 ⇒ 00:03:09.000 Casie Aviles: Okay.
26 00:03:09.000 ⇒ 00:03:12.589 Amber Lin: Oh, I know… an update there, too.
27 00:03:13.080 ⇒ 00:03:14.200 Casie Aviles: Yeah, yeah, yeah.
28 00:03:14.540 ⇒ 00:03:16.569 Amber Lin: How long would each of these take?
29 00:03:18.390 ⇒ 00:03:22.789 Casie Aviles: Yeah, so for the spike, I estimated it as 1…
30 00:03:23.600 ⇒ 00:03:30.040 Casie Aviles: And then for the daily Andy reminders, I would…
31 00:03:30.530 ⇒ 00:03:35.270 Casie Aviles: I think there might be a lot of, other steps that we need to do, so…
32 00:03:36.160 ⇒ 00:03:38.429 Casie Aviles: Let’s just put this to 3 for now.
33 00:03:41.500 ⇒ 00:03:46.060 Casie Aviles: Or, like, yeah… I might be overthinking, but…
34 00:03:46.800 ⇒ 00:03:50.820 Casie Aviles: For this ticket, at least, for the daily and the stuff.
35 00:03:52.680 ⇒ 00:03:55.449 Casie Aviles: Because I think we still need to con-
36 00:03:55.880 ⇒ 00:03:58.909 Casie Aviles: Get, like, the copy and confirm that.
37 00:03:59.250 ⇒ 00:04:00.200 Casie Aviles: And…
38 00:04:02.690 ⇒ 00:04:12.450 Amber Lin: Yeah, I can get all the copies and confirm. I guess on the technical side, is… is this…
39 00:04:13.840 ⇒ 00:04:20.660 Casie Aviles: Yeah, so for… being able to have the live and the send the messages,
40 00:04:21.470 ⇒ 00:04:24.860 Casie Aviles: I need to confirm if,
41 00:04:25.130 ⇒ 00:04:33.050 Casie Aviles: if we really need the service account. If not, if yes, then I will ask Tim for that, so we could have…
42 00:04:33.740 ⇒ 00:04:37.179 Casie Aviles: I guess I might have overlooked something for the setup.
43 00:04:37.910 ⇒ 00:04:40.920 Casie Aviles: So, I’ll do, like, an additional quick…
44 00:04:41.350 ⇒ 00:04:44.400 Casie Aviles: Investigation. It should be just a quick one.
45 00:04:44.770 ⇒ 00:04:50.979 Amber Lin: Gotcha, okay. So determine if we need sparious…
46 00:04:51.370 ⇒ 00:05:03.449 Amber Lin: account, yeah. Let me know, I will confirm who and what to send. Just let me know if we need to escalate, because I know they respond slower. If we can figure it…
47 00:05:04.090 ⇒ 00:05:08.299 Amber Lin: If it’s just figuring out if we need a service account, is that 3 points?
48 00:05:09.220 ⇒ 00:05:17.119 Casie Aviles: It’s going to be, I guess, faster, just… I just have to check it then.
49 00:05:17.780 ⇒ 00:05:19.030 Amber Lin: Okay.
50 00:05:19.290 ⇒ 00:05:22.199 Amber Lin: I would say the long cruise…
51 00:05:22.580 ⇒ 00:05:29.850 Amber Lin: is also very important. How long would that take to add the long crews to the zip code database?
52 00:05:30.980 ⇒ 00:05:34.499 Casie Aviles: I think this will take, like, 2 points for me.
53 00:05:34.500 ⇒ 00:05:35.080 Amber Lin: Okay.
54 00:05:35.210 ⇒ 00:05:36.010 Amber Lin: Okay.
55 00:05:38.250 ⇒ 00:05:47.940 Casie Aviles: And then, which ones are, like… so these two tickets, right, 1138 and 1137 are the most priority… highest priority?
56 00:05:48.510 ⇒ 00:05:51.340 Amber Lin: Yeah, I’ll put this as high. Yeah.
57 00:05:51.510 ⇒ 00:05:57.689 Amber Lin: Mostly because they asked about this today again, and we promised them this week, so I just want to make sure.
58 00:05:57.800 ⇒ 00:06:06.350 Amber Lin: And these two, I will… I’ll move this to… Me…
59 00:06:07.530 ⇒ 00:06:10.140 Amber Lin: And then, here, I think this would be cleaner.
60 00:06:10.690 ⇒ 00:06:18.780 Amber Lin: And then… Yeah, Sam, I wanted to ask about… This one.
61 00:06:20.430 ⇒ 00:06:28.940 Samuel Roberts: Okay, yeah, I wanted to ask about this too, because I… I guess I missed this last week, when… and… and Mustafa was saying that we have the embeddings already in a table?
62 00:06:30.680 ⇒ 00:06:33.700 Amber Lin: What do you mean by embeddings?
63 00:06:34.330 ⇒ 00:06:38.889 Samuel Roberts: Oh, sorry, so, like, in order to search the stuff, the central docs get converted into, like.
64 00:06:39.590 ⇒ 00:06:44.250 Samuel Roberts: Stuff that the AI can read, the embeddings are just numbers. But I guess…
65 00:06:45.440 ⇒ 00:06:49.719 Samuel Roberts: I don’t have access to two of those three links, anyway.
66 00:06:51.240 ⇒ 00:06:59.209 Amber Lin: Do you have, Let’s see… do you have access to the Brainforge ABC account?
67 00:07:02.530 ⇒ 00:07:04.320 Amber Lin: It should be a one pass.
68 00:07:04.320 ⇒ 00:07:04.870 Casie Aviles: I think…
69 00:07:04.870 ⇒ 00:07:05.509 Samuel Roberts: Which one is that?
70 00:07:05.510 ⇒ 00:07:07.650 Casie Aviles: I’ve been having trouble accessing.
71 00:07:07.650 ⇒ 00:07:09.630 Amber Lin: Oh, I see, I see.
72 00:07:10.040 ⇒ 00:07:16.020 Amber Lin: So basically what we have is a Google Drive with their…
73 00:07:16.240 ⇒ 00:07:25.330 Amber Lin: with their different folders and then the different Google Docs in them. I’ve made a list of what we need to add and what we will skip. Okay.
74 00:07:25.460 ⇒ 00:07:29.219 Amber Lin: Actually, these, we should skip the folders, too.
75 00:07:30.230 ⇒ 00:07:30.950 Samuel Roberts: Okay.
76 00:07:31.940 ⇒ 00:07:46.399 Amber Lin: And then… And then I can also write out the order in which we add things, but…
77 00:07:46.600 ⇒ 00:07:49.240 Amber Lin: Essentially, we just want to add these.
78 00:07:50.220 ⇒ 00:07:52.090 Amber Lin: to one document.
79 00:07:52.650 ⇒ 00:08:04.529 Amber Lin: And ideally, we want to format it like our other central docs, which means, it has the content and then a heading, so we can create a table of contents.
80 00:08:05.170 ⇒ 00:08:08.479 Amber Lin: So it would just be a heading, and then the contents underneath.
81 00:08:08.730 ⇒ 00:08:13.640 Amber Lin: Is this possible to have a workflow that automates adding Google Docs?
82 00:08:14.040 ⇒ 00:08:17.330 Amber Lin: to one main Google Doc?
83 00:08:21.220 ⇒ 00:08:25.459 Samuel Roberts: That… I mean, I’m… Let me think.
84 00:08:29.050 ⇒ 00:08:34.770 Samuel Roberts: I’m not… intimately familiar with how that would work. I mean…
85 00:08:34.890 ⇒ 00:08:38.500 Samuel Roberts: Are we trying to keep these up-to-date as well, or is this a one-time thing?
86 00:08:38.870 ⇒ 00:08:46.590 Amber Lin: Just a one-time thing. I’ve been… I used to just manually copy them over, because there wasn’t this many documents.
87 00:08:46.590 ⇒ 00:08:46.950 Samuel Roberts: Okay.
88 00:08:46.950 ⇒ 00:09:02.990 Amber Lin: We really can’t figure out… figure it out this week. We will have to manually copy them, which is okay, because, like, this is the last customer service department, but if we expand in the future and we get more departments, there will be additional docs that will have to do this again.
89 00:09:03.500 ⇒ 00:09:04.940 Samuel Roberts: Okay, there’s…
90 00:09:06.200 ⇒ 00:09:11.520 Samuel Roberts: I don’t wanna… I don’t wanna make too many assumptions, but I’m sure there’s a way to get the content from each one of those.
91 00:09:12.790 ⇒ 00:09:19.039 Samuel Roberts: Probably a way to… yeah, I mean, have you guys used… you guys might have more intimate familiarity with
92 00:09:19.480 ⇒ 00:09:23.589 Samuel Roberts: the Google, like, APIs here,
93 00:09:25.560 ⇒ 00:09:28.469 Samuel Roberts: How have we done that in the past, accessing these documents?
94 00:09:28.810 ⇒ 00:09:30.930 Samuel Roberts: Or actually fanny documents, I guess.
95 00:09:34.340 ⇒ 00:09:36.730 Mustafa Raja: Accessing them in any time or something.
96 00:09:37.260 ⇒ 00:09:39.090 Samuel Roberts: Well, or just in general, like.
97 00:09:39.220 ⇒ 00:09:46.289 Samuel Roberts: I don’t have a ton of familiarity with, like, how to programmatically access these documents and copy and paste into a new document.
98 00:09:46.490 ⇒ 00:09:51.030 Samuel Roberts: I’m sure there’s a way to do it, I just don’t know how in-depth it’s gonna… how, like, technical it’s gonna have to get.
99 00:09:52.040 ⇒ 00:09:54.760 Casie Aviles: Yeah, probably it will involve, like…
100 00:09:55.490 ⇒ 00:10:00.030 Casie Aviles: Google API, Google Drive API, right? Or the Google Docs API.
101 00:10:00.610 ⇒ 00:10:04.790 Samuel Roberts: Right, and so I imagine… have you guys used that at all? I just don’t know what the familiarity of the team in general is with that.
102 00:10:04.790 ⇒ 00:10:12.890 Mustafa Raja: Yes, so, we have, Google Docs, what’s it called, Node in NITN?
103 00:10:12.990 ⇒ 00:10:26.989 Mustafa Raja: And that works pretty good. There’s two ways we can output the data. One is JSON. It’s very ugly, and one is, they call it a simplified version.
104 00:10:27.050 ⇒ 00:10:38.819 Mustafa Raja: Which would have all of the text only. So, yeah, that is how we can get, if it’s Google Docs, we can get the textual data.
105 00:10:38.900 ⇒ 00:10:41.689 Mustafa Raja: From the dock itself.
106 00:10:41.690 ⇒ 00:10:44.810 Samuel Roberts: What do these documents look like? Like, are they just.
107 00:10:44.810 ⇒ 00:10:48.870 Amber Lin: What do you… Yeah, I can share screen. I think…
108 00:10:48.980 ⇒ 00:10:54.779 Samuel Roberts: Yeah. We… remember when we first did the Pest Central doc? I think Miguel also ran the workflow.
109 00:10:54.780 ⇒ 00:10:55.379 Amber Lin: to have…
110 00:10:55.380 ⇒ 00:10:55.840 Samuel Roberts: Okay.
111 00:10:55.840 ⇒ 00:11:03.129 Amber Lin: But we ran into the same problem, that we only… got… Text, and there.
112 00:11:03.130 ⇒ 00:11:03.740 Samuel Roberts: Okay.
113 00:11:03.740 ⇒ 00:11:05.769 Amber Lin: The formatting was really hard, we weren’t.
114 00:11:05.770 ⇒ 00:11:06.500 Casie Aviles: We’re able to get.
115 00:11:06.790 ⇒ 00:11:07.860 Amber Lin: highlights.
116 00:11:07.860 ⇒ 00:11:08.660 Samuel Roberts: I see, okay.
117 00:11:08.660 ⇒ 00:11:14.110 Amber Lin: Probably… we’ll probably end up doing it manually, like, we’ll just split the work, and then we’ll just copy and.
118 00:11:14.110 ⇒ 00:11:17.099 Samuel Roberts: How many are there? How many we’re talking about here?
119 00:11:17.400 ⇒ 00:11:18.030 Amber Lin: Like, it’s…
120 00:11:18.030 ⇒ 00:11:20.310 Samuel Roberts: Oh, no, no, no, okay. No, no.
121 00:11:20.310 ⇒ 00:11:22.920 Amber Lin: No, it just kept growing.
122 00:11:22.920 ⇒ 00:11:28.789 Samuel Roberts: Yeah, yeah, okay, no, no, no, I can… I… okay, so we’ve done this with NIDN, and it wasn’t ideal before, is that what I’m hearing?
123 00:11:30.420 ⇒ 00:11:32.309 Mustafa Raja: Yeah. Miguel had a workflow there.
124 00:11:32.310 ⇒ 00:11:33.009 Samuel Roberts: Copy just the text.
125 00:11:33.010 ⇒ 00:11:41.729 Mustafa Raja: But if we do the extended export from Google Docs, that is going to have proper… what’s it called?
126 00:11:42.140 ⇒ 00:11:43.319 Samuel Roberts: That’s what I’m wondering, can we format that right?
127 00:11:43.320 ⇒ 00:11:44.020 Mustafa Raja: Into another one?
128 00:11:44.800 ⇒ 00:11:45.940 Samuel Roberts: NNN?
129 00:11:48.380 ⇒ 00:11:57.950 Mustafa Raja: The insert is weird. For NLT and the insert, I don’t think we can insert using JSON. We’ll have to use, Google Docs API.
130 00:11:58.600 ⇒ 00:12:06.940 Samuel Roberts: Okay. Yeah, okay, now that I… sorry, I didn’t have a ton of context for this, and I just… I didn’t really understand. I think I get it now. If I can…
131 00:12:07.550 ⇒ 00:12:12.099 Samuel Roberts: get access so that, the… what is it, Triandeater one?
132 00:12:12.890 ⇒ 00:12:13.560 Samuel Roberts: Where is that?
133 00:12:13.560 ⇒ 00:12:17.770 Amber Lin: It’s the… Brain Forge Echo, and Eater.
134 00:12:17.770 ⇒ 00:12:19.170 Samuel Roberts: Yeah, okay.
135 00:12:20.340 ⇒ 00:12:23.200 Samuel Roberts: I’m gonna try again. I definitely had issues before.
136 00:12:23.200 ⇒ 00:12:23.620 Amber Lin: signing.
137 00:12:23.620 ⇒ 00:12:26.609 Samuel Roberts: Yeah, couldn’t verify, it wants me to recover again.
138 00:12:27.550 ⇒ 00:12:29.670 Samuel Roberts: Okay.
139 00:12:36.160 ⇒ 00:12:39.539 Samuel Roberts: Yeah, I definitely… I still couldn’t sign you in, so I don’t know.
140 00:12:39.650 ⇒ 00:12:41.050 Samuel Roberts: Okay.
141 00:12:42.730 ⇒ 00:12:49.199 Amber Lin: If you can’t sign in, maybe I could reassign this to Mustafa, and you guys can pair on how it.
142 00:12:49.200 ⇒ 00:12:51.129 Samuel Roberts: We can definitely pair on it, yeah.
143 00:12:51.130 ⇒ 00:12:51.850 Amber Lin: Okay.
144 00:12:53.030 ⇒ 00:13:00.740 Mustafa Raja: So the goal is that we maintain the… Formats while adding the… Oh, no.
145 00:13:00.740 ⇒ 00:13:10.110 Amber Lin: Yeah, ideally, because if not, we will have to reformat it, or they might not understand what we have copied over.
146 00:13:10.110 ⇒ 00:13:10.920 Mustafa Raja: over.
147 00:13:11.390 ⇒ 00:13:14.509 Amber Lin: Yeah, especially the bullet points and…
148 00:13:14.510 ⇒ 00:13:15.230 Samuel Roberts: Right, right.
149 00:13:15.230 ⇒ 00:13:15.710 Amber Lin: and all.
150 00:13:15.710 ⇒ 00:13:17.900 Samuel Roberts: And then the, the final, just…
151 00:13:18.150 ⇒ 00:13:24.070 Samuel Roberts: Maybe, like I said, a little out of context. So this is… we’re putting together a new central doc for…
152 00:13:25.760 ⇒ 00:13:27.580 Amber Lin: Yes, so we’re combining…
153 00:13:27.890 ⇒ 00:13:28.529 Samuel Roberts: Combining a whole bunch.
154 00:13:28.530 ⇒ 00:13:37.449 Amber Lin: for… yeah, because these are all the SOPs and information that the commercial department uses, and they.
155 00:13:37.450 ⇒ 00:13:38.100 Samuel Roberts: Got it.
156 00:13:38.100 ⇒ 00:13:57.540 Amber Lin: The commercial department covers commercial pests, and then there’s commercial lawn, commercial, just general sales, and they put it under the same department, which is commercial, and they are able to… they need access to all of these to schedule and answer the questions.
157 00:13:58.060 ⇒ 00:14:00.169 Samuel Roberts: So then Andy’s gonna be looking at this eventually?
158 00:14:00.690 ⇒ 00:14:01.550 Amber Lin: Yeah.
159 00:14:01.990 ⇒ 00:14:02.660 Samuel Roberts: Okay.
160 00:14:07.390 ⇒ 00:14:11.440 Samuel Roberts: Okay, so is the… I guess, if Andy’s gonna be looking at this.
161 00:14:13.310 ⇒ 00:14:19.660 Samuel Roberts: Is Andy… Andy’s just pulling from the… Mustafa and Casey, the embeddings that are in the…
162 00:14:20.590 ⇒ 00:14:21.180 Mustafa Raja: Province.
163 00:14:21.180 ⇒ 00:14:24.790 Samuel Roberts: In Superbase, and then just returning text from there, right?
164 00:14:24.790 ⇒ 00:14:25.160 Casie Aviles: Yeah.
165 00:14:25.160 ⇒ 00:14:25.660 Mustafa Raja: Yes.
166 00:14:26.010 ⇒ 00:14:29.300 Samuel Roberts: So, we’re gonna lose all this formatting anyway, right?
167 00:14:29.300 ⇒ 00:14:33.620 Amber Lin: You’re right. What if we… like, what if we try a new approach and.
168 00:14:33.620 ⇒ 00:14:35.190 Samuel Roberts: That’s what I’m wondering, I think I’m…
169 00:14:35.190 ⇒ 00:14:36.350 Amber Lin: Beddings?
170 00:14:37.690 ⇒ 00:14:43.270 Samuel Roberts: Well, that’s what I’m thinking, if we just needed to create embeddings for all of these, we don’t need an actual Google Doc that has it all.
171 00:14:45.710 ⇒ 00:14:53.480 Samuel Roberts: We just would need to ingest each one of these documents, which would be a different process than copying and pasting into every… every one into one document.
172 00:14:53.480 ⇒ 00:14:54.759 Mustafa Raja: My aunt.
173 00:14:55.320 ⇒ 00:14:56.120 Samuel Roberts: Go ahead.
174 00:14:56.290 ⇒ 00:15:04.620 Mustafa Raja: My understanding of having the formatting, is only that, ABC people also look into it.
175 00:15:06.100 ⇒ 00:15:08.210 Samuel Roberts: Will they be… That’s what I’m asking.
176 00:15:08.380 ⇒ 00:15:11.709 Amber Lin: Yeah, in the mor… this is why I think you were…
177 00:15:11.970 ⇒ 00:15:16.840 Amber Lin: suggestion might be very viable, because in the morning, when I talk to them.
178 00:15:16.860 ⇒ 00:15:29.400 Amber Lin: Ideally, we want them to all use ANDI, and sometimes the problem is that they use our central dock instead of using ANDI, because the central dock is also easier than this current system.
179 00:15:29.400 ⇒ 00:15:42.819 Amber Lin: So, what you propose is it might actually work, because there will be no central dock, and then they will either have to use this cranky system or use ANDI, but we risk that they never use ANDI.
180 00:15:43.150 ⇒ 00:15:45.559 Samuel Roberts: Exactly, that’s what I’m a little worried about.
181 00:15:45.930 ⇒ 00:15:52.690 Amber Lin: But maybe it’s the same, because it’s, like, Andy or the Central Doc is a new thing, so maybe.
182 00:15:52.690 ⇒ 00:15:53.470 Samuel Roberts: It’s easier to.
183 00:15:53.470 ⇒ 00:16:05.670 Amber Lin: convert them to one thing instead of convert them to two things. I just want to know, like, how… how reliable it is if we’re ingesting so many documents, and then having to
184 00:16:05.900 ⇒ 00:16:09.149 Amber Lin: If we do that, we need to maintain the updates here.
185 00:16:10.140 ⇒ 00:16:16.900 Amber Lin: So, each time they update each and every single one of these documents, we need to change our…
186 00:16:17.540 ⇒ 00:16:20.350 Amber Lin: whatever, the embeddings were whatever’s in Superbase.
187 00:16:20.350 ⇒ 00:16:20.890 Samuel Roberts: Yeah.
188 00:16:20.890 ⇒ 00:16:21.650 Amber Lin: That will be…
189 00:16:21.650 ⇒ 00:16:27.430 Samuel Roberts: Well, we’d have… we’d have to do that anyway. So, the workflow would have to be, they make a change to a document.
190 00:16:30.250 ⇒ 00:16:30.830 Samuel Roberts: Nightmare.
191 00:16:30.830 ⇒ 00:16:40.599 Amber Lin: Right now, they make the change to the central doc, so we only listen to one document, well, for each department.
192 00:16:40.860 ⇒ 00:16:47.540 Amber Lin: But if we… In just… if we don’t create a central dog, then we have to listen to
193 00:16:47.680 ⇒ 00:16:52.860 Amber Lin: the changes in, say, these 180 documents. That’s a trade-off.
194 00:16:53.890 ⇒ 00:16:58.599 Samuel Roberts: Hmm… Oh, okay.
195 00:17:01.240 ⇒ 00:17:06.349 Mustafa Raja: We can have a schedule… schedule of the week or something, if you want to avoid this.
196 00:17:06.739 ⇒ 00:17:08.329 Samuel Roberts: That’s a good point. We could just…
197 00:17:10.239 ⇒ 00:17:11.919 Samuel Roberts: See if there have been changes every day.
198 00:17:11.920 ⇒ 00:17:18.390 Mustafa Raja: Because the other day, Utam said that, we don’t have it… we don’t… we don’t really need to have it.
199 00:17:18.740 ⇒ 00:17:21.790 Mustafa Raja: Real time, we can have some sort of facility.
200 00:17:23.540 ⇒ 00:17:35.600 Samuel Roberts: Yeah, he did mention that. The real-time func… the real-time nature of it is a little bit more complex, but if we have a thing that runs at, you know, 3 AM, checks the documents, see if there’s any changes, runs a…
201 00:17:36.110 ⇒ 00:17:37.780 Samuel Roberts: Runs a script where there have been.
202 00:17:37.780 ⇒ 00:17:40.430 Mustafa Raja: Yeah, I think, Amber… Amber, what do you think about this?
203 00:17:41.620 ⇒ 00:17:46.439 Amber Lin: I think it’s viable. I… I don’t think I can sign off on it.
204 00:17:46.570 ⇒ 00:17:48.720 Amber Lin: This is something that we…
205 00:17:48.880 ⇒ 00:17:55.449 Amber Lin: Should probably write up a little bit, look into it a bit more, and maybe the…
206 00:17:55.600 ⇒ 00:18:07.939 Amber Lin: Like, the processing time of listening to all these documents, like, how much are we going to miss? Or, like, if issues happen, we might need to find
207 00:18:08.160 ⇒ 00:18:11.750 Amber Lin: through… go through all the 200 documents, and then.
208 00:18:11.750 ⇒ 00:18:12.100 Casie Aviles: Yeah.
209 00:18:12.100 ⇒ 00:18:14.620 Amber Lin: fix that. Like, that’s a downside.
210 00:18:14.620 ⇒ 00:18:16.530 Samuel Roberts: What are we fixing in documents?
211 00:18:16.830 ⇒ 00:18:20.649 Amber Lin: Like, if something… if, like, a sync goes wrong, or if…
212 00:18:20.650 ⇒ 00:18:21.500 Samuel Roberts: Oh, I see.
213 00:18:21.500 ⇒ 00:18:28.110 Amber Lin: Like, something goes wrong, we… it’s not just in one, but we’ll have to find it, and then…
214 00:18:28.210 ⇒ 00:18:31.640 Amber Lin: Fix that particular one, and that might be an issue.
215 00:18:32.650 ⇒ 00:18:37.250 Samuel Roberts: Is that different than if we were to put them all into one document and have to rerun that every time anyway?
216 00:18:40.860 ⇒ 00:18:54.809 Amber Lin: What do you mean? I think my… I was just saying that there’s too many, many possible chances of error when we have many documents versus one document, and we know that that’s the one causing problems.
217 00:18:54.810 ⇒ 00:19:00.159 Samuel Roberts: So are they… are they then only making changes to that? So the central document becomes the source of truth?
218 00:19:00.620 ⇒ 00:19:01.260 Amber Lin: Yes.
219 00:19:01.880 ⇒ 00:19:04.499 Samuel Roberts: And no one, no one ever looked at these other documents ever again.
220 00:19:05.340 ⇒ 00:19:10.319 Amber Lin: Ideally, that’s what the trainers tell them. The trainers will only update the central dock.
221 00:19:10.940 ⇒ 00:19:13.660 Samuel Roberts: Okay Okay.
222 00:19:14.350 ⇒ 00:19:15.570 Samuel Roberts: I understand that.
223 00:19:16.280 ⇒ 00:19:20.340 Samuel Roberts: Either way, we need to do the embeddings to get it into Andy, so I think the…
224 00:19:21.590 ⇒ 00:19:28.419 Samuel Roberts: Putting together the one doc, if we can come up with a script that can do that cleanly and not lose the formatting.
225 00:19:29.220 ⇒ 00:19:30.460 Samuel Roberts: is…
226 00:19:31.590 ⇒ 00:19:35.289 Samuel Roberts: Worth doing, if that’s the… if it solves that problem.
227 00:19:35.720 ⇒ 00:19:39.670 Samuel Roberts: The embeddings are gonna have to happen anyway.
228 00:19:39.940 ⇒ 00:19:43.680 Samuel Roberts: So, like, whether or not we do it from… each document.
229 00:19:43.850 ⇒ 00:19:46.690 Samuel Roberts: or not, I think it’s fine to…
230 00:19:47.780 ⇒ 00:19:53.590 Samuel Roberts: to do either way. So yeah, okay, we probably do want to put a central dock together then, a commercial central dock.
231 00:19:53.730 ⇒ 00:19:57.159 Samuel Roberts: That is the only thing that we will ever read from.
232 00:19:58.720 ⇒ 00:19:59.520 Samuel Roberts: Right?
233 00:20:00.390 ⇒ 00:20:05.680 Amber Lin: Yeah, is there a spike ticket, some template somewhere?
234 00:20:08.410 ⇒ 00:20:11.289 Samuel Roberts: Somewhere in Notion, I believe there is.
235 00:20:11.290 ⇒ 00:20:12.330 Amber Lin: I see.
236 00:20:12.330 ⇒ 00:20:16.420 Samuel Roberts: Haha, yeah, I… Yeah.
237 00:20:19.030 ⇒ 00:20:20.499 Samuel Roberts: I definitely have done that at some point.
238 00:20:20.500 ⇒ 00:20:21.340 Amber Lin: Sure!
239 00:20:26.010 ⇒ 00:20:40.999 Amber Lin: Commercial, Andy. I will need some help with this. So, embeddings, and then… Central Doc… How do we…
240 00:20:52.470 ⇒ 00:20:55.560 Amber Lin: What’s the question on embeddings? Like, should we…
241 00:20:55.560 ⇒ 00:21:00.270 Samuel Roberts: Actually, hold on, is there… is there any other, like, add-ons that we could find that might do this, you think?
242 00:21:02.030 ⇒ 00:21:05.870 Casie Aviles: You mean, like, a built solution already?
243 00:21:06.040 ⇒ 00:21:07.860 Samuel Roberts: Yeah, like, something that’s not…
244 00:21:08.060 ⇒ 00:21:15.580 Samuel Roberts: Like, something that is just on the Google… like, I don’t know. I just don’t know what’s out there. I don’t know what people have done already. Like, this is where we need to probably do a little bit of the…
245 00:21:15.580 ⇒ 00:21:16.600 Casie Aviles: Yeah…
246 00:21:17.230 ⇒ 00:21:19.100 Samuel Roberts: I mean, the only other…
247 00:21:19.960 ⇒ 00:21:20.949 Casie Aviles: Yeah, yeah.
248 00:21:21.680 ⇒ 00:21:24.380 Samuel Roberts: Not allowed by your administrator, cool, okay.
249 00:21:24.500 ⇒ 00:21:29.480 Samuel Roberts: You know, let’s just… yeah, okay. I can dig into trying to see the…
250 00:21:30.830 ⇒ 00:21:40.850 Samuel Roberts: the API and just get a script, because if it’s only a one-time thing we’re doing, it’s not… I’m less worried about it, being, like, in an event or something that has to run more frequently.
251 00:21:44.250 ⇒ 00:21:45.140 Samuel Roberts: Okay.
252 00:21:49.100 ⇒ 00:21:53.139 Samuel Roberts: How best to do this, though, if I don’t have access to things? That’s the other side of this, right?
253 00:21:53.390 ⇒ 00:21:57.269 Samuel Roberts: So then… Mustafa, you and I may have to pair on this.
254 00:21:57.270 ⇒ 00:21:58.110 Mustafa Raja: Yeah.
255 00:21:59.360 ⇒ 00:21:59.700 Samuel Roberts: Okay.
256 00:22:00.000 ⇒ 00:22:07.300 Mustafa Raja: I’ve looked into it before, when we were, trying to build a master agent, something, for ABC.
257 00:22:09.010 ⇒ 00:22:09.690 Samuel Roberts: Okay.
258 00:22:09.930 ⇒ 00:22:10.530 Mustafa Raja: Yeah.
259 00:22:11.990 ⇒ 00:22:16.140 Samuel Roberts: Alright, let’s set some time, maybe… Peter…
260 00:22:16.140 ⇒ 00:22:18.830 Mustafa Raja: What would be a good… good deadline for this, Amber?
261 00:22:21.130 ⇒ 00:22:25.630 Amber Lin: It’s end of week unreasonable.
262 00:22:29.280 ⇒ 00:22:42.100 Mustafa Raja: I think Sam and I can pair, I guess, tomorrow, because today he’s out of office.
263 00:22:42.620 ⇒ 00:22:46.069 Samuel Roberts: Yeah, I’ll be back by 4PM my time, hopefully, but we’ll see.
264 00:22:46.070 ⇒ 00:22:48.640 Mustafa Raja: I’ll be up, if you want to sing then.
265 00:22:49.170 ⇒ 00:22:49.960 Samuel Roberts: Okay.
266 00:22:50.550 ⇒ 00:22:59.999 Amber Lin: There’s also the option of doing it manually, copy and pasting, because, right, a lot of it is skipped, we probably have, like, a hundred.
267 00:23:00.560 ⇒ 00:23:01.860 Amber Lin: Documents…
268 00:23:02.500 ⇒ 00:23:04.879 Amber Lin: That we do need to add, like…
269 00:23:05.250 ⇒ 00:23:11.980 Amber Lin: If… if this will take very long, it doesn’t… or if it doesn’t seem that plausible,
270 00:23:12.130 ⇒ 00:23:16.450 Amber Lin: Yosefa, you and I can just split the work, and we can just copy and paste, and…
271 00:23:16.450 ⇒ 00:23:23.319 Mustafa Raja: Okay, yeah, I guess then, I guess then Sam and I will pair, and see if this is.
272 00:23:23.320 ⇒ 00:23:24.000 Samuel Roberts: Yeah, we can.
273 00:23:24.000 ⇒ 00:23:24.400 Mustafa Raja: Or no.
274 00:23:24.400 ⇒ 00:23:25.780 Samuel Roberts: Hope we’ll figure that out tonight.
275 00:23:25.960 ⇒ 00:23:27.370 Mustafa Raja: Yeah, okay.
276 00:23:27.370 ⇒ 00:23:29.050 Samuel Roberts: Able to do it. Yeah.
277 00:23:29.050 ⇒ 00:23:35.830 Mustafa Raja: by end of day, and then, let team know in, ABC channel.
278 00:23:36.500 ⇒ 00:23:39.089 Mustafa Raja: So we can make appropriate decisions.
279 00:23:39.860 ⇒ 00:23:41.070 Amber Lin: Okay, sounds good.
280 00:23:41.270 ⇒ 00:23:42.060 Amber Lin: Great.
281 00:23:42.480 ⇒ 00:23:46.780 Amber Lin: Okay, I think that’s the… that’s the main thing. So, Casey.
282 00:23:46.780 ⇒ 00:23:47.230 Samuel Roberts: Okay.
283 00:23:47.230 ⇒ 00:23:50.119 Amber Lin: work, and we have our work. Great.
284 00:23:51.460 ⇒ 00:23:52.220 Amber Lin: Alright.
285 00:23:52.510 ⇒ 00:23:53.489 Mustafa Raja: Cool. Thank you.
286 00:23:53.740 ⇒ 00:23:54.950 Amber Lin: Alright, thanks, team.
287 00:23:54.950 ⇒ 00:23:55.970 Samuel Roberts: Alrighty, yep.
288 00:23:55.970 ⇒ 00:23:56.440 Casie Aviles: Thank you.
289 00:23:56.440 ⇒ 00:23:57.000 Amber Lin: Bye.
290 00:23:57.000 ⇒ 00:23:57.440 Mustafa Raja: Dang back.