Meeting Title: Brainforge AI Weekly Planning Date: 2026-02-11 Meeting participants: Mustafa Raja, Casie Aviles, Amber Lin, Samuel Roberts
WEBVTT
1 00:01:36.220 ⇒ 00:01:37.460 Casie Aviles: Hey, Mustafa.
2 00:01:40.850 ⇒ 00:01:42.119 Mustafa Raja: Hey, how are you?
3 00:01:43.910 ⇒ 00:01:45.060 Casie Aviles: Yeah, doing good.
4 00:01:46.630 ⇒ 00:01:53.570 Casie Aviles: I guess, yeah, let’s just wait a few more minutes for… The others, if they’re… Coming.
5 00:01:54.260 ⇒ 00:01:57.710 Mustafa Raja: Yeah… I think Sam, Sam is coming, definitely.
6 00:02:04.250 ⇒ 00:02:04.770 Amber Lin: Hello!
7 00:02:04.770 ⇒ 00:02:05.800 Mustafa Raja: How’s the name?
8 00:02:07.760 ⇒ 00:02:09.880 Casie Aviles: Hey. Yeah, doing good.
9 00:02:17.420 ⇒ 00:02:21.569 Amber Lin: Okay. Do you guys know if CM is coming?
10 00:02:22.120 ⇒ 00:02:25.050 Mustafa Raja: Yes, Sam is coming in a few moments.
11 00:02:26.820 ⇒ 00:02:28.490 Casie Aviles: Yeah, he’s just grabbing coffee.
12 00:02:29.040 ⇒ 00:02:29.610 Amber Lin: Okay.
13 00:02:30.500 ⇒ 00:02:39.570 Amber Lin: Sounds good. I can pull up our Gantt to check where we’re at.
14 00:02:41.800 ⇒ 00:02:44.879 Mustafa Raja: We just got unblocked on a Cloud SQL.
15 00:02:45.110 ⇒ 00:03:00.090 Amber Lin: Oh, wow, awesome. Okay. Let me… 2… this, so… We are… Shane R… Let’s see, where are we?
16 00:03:01.520 ⇒ 00:03:07.550 Amber Lin: This week… We are… Claude.
17 00:03:08.450 ⇒ 00:03:09.680 Amber Lin: That’s cool.
18 00:03:11.260 ⇒ 00:03:13.039 Mustafa Raja: Yes, we can now set this up, yeah.
19 00:03:13.710 ⇒ 00:03:15.400 Amber Lin: Oh, oh, what happened?
20 00:03:15.880 ⇒ 00:03:17.230 Mustafa Raja: It collapsed.
21 00:03:18.040 ⇒ 00:03:19.280 Amber Lin: Okay.
22 00:03:19.410 ⇒ 00:03:24.409 Amber Lin: So, I will expand this… To the end of this week?
23 00:03:25.990 ⇒ 00:03:29.890 Casie Aviles: Sure, I’ve… I’m currently at home improvement now.
24 00:03:32.020 ⇒ 00:03:37.809 Casie Aviles: So, we… I’ve added, mechanical and blood.
25 00:03:39.070 ⇒ 00:03:42.880 Amber Lin: Awesome, so I’ll say this is… Seriously.
26 00:03:45.050 ⇒ 00:03:46.649 Mustafa Raja: We have to mark it as an example.
27 00:03:46.650 ⇒ 00:03:47.320 Amber Lin: Sweet.
28 00:03:47.590 ⇒ 00:03:50.499 Amber Lin: This we have started. Hello!
29 00:03:51.450 ⇒ 00:03:52.700 Amber Lin: So…
30 00:03:56.640 ⇒ 00:03:58.200 Casie Aviles: I must say…
31 00:03:59.100 ⇒ 00:04:05.800 Casie Aviles: Also, I’ll just note that I’ll probably just focus on adding the zips. I may not be able to look
32 00:04:05.950 ⇒ 00:04:09.819 Casie Aviles: Work on any migration, since the validation is…
33 00:04:10.020 ⇒ 00:04:14.719 Casie Aviles: Taking a while for me, but I wanted to make sure that
34 00:04:15.390 ⇒ 00:04:22.220 Casie Aviles: You know, to eliminate the… the factor that… There’s any missing…
35 00:04:22.490 ⇒ 00:04:25.000 Amber Lin: Yeah. Makes sense. Zips.
36 00:04:25.290 ⇒ 00:04:37.129 Amber Lin: Sounds like it will take at least another week, right? Should I… can I expand this? Because there’s also the pest tax and all that. I don’t think you should cram that in this week, I’ll be too much work.
37 00:04:38.330 ⇒ 00:04:39.869 Casie Aviles: Okay, yeah, we can do that.
38 00:04:40.200 ⇒ 00:04:40.760 Amber Lin: Okay.
39 00:04:40.760 ⇒ 00:04:41.160 Samuel Roberts: what…
40 00:04:41.160 ⇒ 00:04:42.330 Amber Lin: Huh.
41 00:04:42.330 ⇒ 00:04:44.959 Samuel Roberts: migration tasks that you have is someone I can grab?
42 00:04:46.930 ⇒ 00:04:51.339 Casie Aviles: Let’s see… I think it’s mostly related.
43 00:04:51.340 ⇒ 00:04:54.720 Amber Lin: This week… this is this… this one.
44 00:04:54.720 ⇒ 00:04:55.160 Samuel Roberts: Mmm.
45 00:04:56.350 ⇒ 00:05:00.189 Amber Lin: And then, yeah, I think it’s just that, and Mustaf was doing the…
46 00:05:00.410 ⇒ 00:05:03.630 Amber Lin: And then Migrate AI models, which I don’t know who it is doing.
47 00:05:04.550 ⇒ 00:05:11.270 Amber Lin: Oh, and the text to SQL accuracy, which I think we should do after we have the zips in there.
48 00:05:11.680 ⇒ 00:05:12.310 Samuel Roberts: Yeah, definitely.
49 00:05:13.010 ⇒ 00:05:13.690 Amber Lin: Okay.
50 00:05:20.060 ⇒ 00:05:23.600 Amber Lin: How do I move that depend… oh…
51 00:05:28.470 ⇒ 00:05:29.950 Amber Lin: Oh, man.
52 00:05:31.260 ⇒ 00:05:33.770 Amber Lin: How do I delete this dependency?
53 00:05:33.770 ⇒ 00:05:37.379 Mustafa Raja: Oh, if you… if you would right-click that, right-click that.
54 00:05:37.380 ⇒ 00:05:40.500 Amber Lin: Oh, okay, let me delete.
55 00:05:40.630 ⇒ 00:05:41.620 Amber Lin: Close.
56 00:05:43.060 ⇒ 00:05:43.930 Amber Lin: Great.
57 00:05:44.610 ⇒ 00:05:45.720 Amber Lin: So…
58 00:05:48.840 ⇒ 00:05:52.040 Amber Lin: Oh… Awful.
59 00:06:04.210 ⇒ 00:06:09.170 Amber Lin: Cool. Okay, so that gets… Push back another week.
60 00:06:15.000 ⇒ 00:06:18.120 Amber Lin: And then… so what else?
61 00:06:18.580 ⇒ 00:06:20.480 Amber Lin: is on Casey’s plate.
62 00:06:21.940 ⇒ 00:06:30.350 Amber Lin: This… Classiol, in progress. Yeah, the menstrual locks to be cute. Are we doing that this week, or…
63 00:06:30.650 ⇒ 00:06:34.489 Mustafa Raja: Yeah, that depends on Cloud SQL, right? So…
64 00:06:34.840 ⇒ 00:06:38.250 Mustafa Raja: Set that up, and then… and then it shouldn’t be much.
65 00:06:38.780 ⇒ 00:06:41.290 Mustafa Raja: Much of the work. I just…
66 00:06:41.290 ⇒ 00:06:43.819 Amber Lin: So, is this gonna be next week, or can we.
67 00:06:43.820 ⇒ 00:06:48.740 Mustafa Raja: I think, yeah, I think, I think I should be able to take this this week, right?
68 00:06:48.740 ⇒ 00:06:50.870 Amber Lin: Okay, so I’ll keep it there.
69 00:06:50.870 ⇒ 00:06:51.460 Mustafa Raja: Yeah.
70 00:06:51.460 ⇒ 00:06:52.190 Amber Lin: Mmm…
71 00:06:52.190 ⇒ 00:06:58.870 Mustafa Raja: I’m just going to initialize the Cloud SQL today, and then my team will take on master logs to be queued today.
72 00:07:00.400 ⇒ 00:07:01.310 Amber Lin: Hmm, okay.
73 00:07:01.490 ⇒ 00:07:02.470 Amber Lin: Sounds good.
74 00:07:02.940 ⇒ 00:07:04.030 Mustafa Raja: So…
75 00:07:04.030 ⇒ 00:07:09.070 Amber Lin: What about Andy’s staging? Are you gonna do that this week, or…
76 00:07:11.310 ⇒ 00:07:14.399 Mustafa Raja: I think this is already established, no?
77 00:07:14.400 ⇒ 00:07:16.240 Amber Lin: Oh, I thought we… I know we have…
78 00:07:16.240 ⇒ 00:07:16.730 Casie Aviles: develop.
79 00:07:16.730 ⇒ 00:07:17.460 Amber Lin: development.
80 00:07:17.460 ⇒ 00:07:17.920 Mustafa Raja: Oh.
81 00:07:18.200 ⇒ 00:07:20.330 Mustafa Raja: Do we need this?
82 00:07:23.380 ⇒ 00:07:24.199 Casie Aviles: Yeah, it was a pleasure.
83 00:07:24.200 ⇒ 00:07:24.820 Amber Lin: you.
84 00:07:25.000 ⇒ 00:07:26.100 Amber Lin: Or, for you.
85 00:07:26.100 ⇒ 00:07:26.760 Samuel Roberts: Right.
86 00:07:27.720 ⇒ 00:07:30.799 Mustafa Raja: Sorry, I was responding to a message from Utam, what’s up?
87 00:07:31.060 ⇒ 00:07:31.720 Mustafa Raja: Okay, do we…
88 00:07:31.720 ⇒ 00:07:33.450 Amber Lin: I need anti-staging this week.
89 00:07:33.450 ⇒ 00:07:35.579 Mustafa Raja: We already have development.
90 00:07:35.580 ⇒ 00:07:40.710 Samuel Roberts: I don’t… think we necessarily need it this week, if we can test with development.
91 00:07:40.710 ⇒ 00:07:41.400 Mustafa Raja: Yeah.
92 00:07:43.290 ⇒ 00:07:45.819 Samuel Roberts: Staging’s kind of for them, and I don’t know if we’re there yet.
93 00:07:46.320 ⇒ 00:07:46.930 Amber Lin: Okay.
94 00:07:47.240 ⇒ 00:07:52.750 Amber Lin: So, let’s see… do we need to migrate AI models this week?
95 00:07:53.840 ⇒ 00:07:55.379 Samuel Roberts: Is that the…
96 00:07:56.410 ⇒ 00:08:02.250 Mustafa Raja: I think, and this is, this is, for them to host the AI models, right?
97 00:08:02.510 ⇒ 00:08:04.430 Samuel Roberts: Oh, right, right, right. Okay.
98 00:08:05.070 ⇒ 00:08:08.900 Mustafa Raja: Because we’re using our internal… internal models.
99 00:08:09.280 ⇒ 00:08:13.360 Samuel Roberts: Yeah, did we… do we have a plan for that?
100 00:08:15.390 ⇒ 00:08:16.159 Mustafa Raja: I don’t know.
101 00:08:16.630 ⇒ 00:08:17.180 Samuel Roberts: Yeah.
102 00:08:17.230 ⇒ 00:08:19.419 Mustafa Raja: I think we just need to communicate with them, right?
103 00:08:20.190 ⇒ 00:08:21.300 Samuel Roberts: Yeah…
104 00:08:21.760 ⇒ 00:08:22.090 Amber Lin: Okay.
105 00:08:22.090 ⇒ 00:08:22.620 Casie Aviles: Yeah, we don’.
106 00:08:22.620 ⇒ 00:08:24.079 Amber Lin: I mean, we can start the plan.
107 00:08:24.080 ⇒ 00:08:25.699 Mustafa Raja: I have some context on that?
108 00:08:27.500 ⇒ 00:08:33.360 Casie Aviles: Yeah, we just have, like, the old migration plan from December. Okay.
109 00:08:33.809 ⇒ 00:08:43.530 Casie Aviles: And yeah, like, what we did just talk about is they’re open to, like, moving it to, like, getting their own OpenAI platform if we wanted their…
110 00:08:43.539 ⇒ 00:08:48.109 Samuel Roberts: Okay, that’s what I was wondering. I didn’t know if this was gonna have to be a GCP thing or not, but…
111 00:08:48.299 ⇒ 00:08:49.819 Samuel Roberts: I wasn’t sure how that would work.
112 00:08:53.670 ⇒ 00:08:57.089 Mustafa Raja: I think OpenAI is retiring 4-0, right?
113 00:08:57.390 ⇒ 00:09:04.050 Samuel Roberts: Alright… So we may have to do some testing of some other models anyway.
114 00:09:04.300 ⇒ 00:09:05.100 Mustafa Raja: Yeah.
115 00:09:06.040 ⇒ 00:09:17.729 Amber Lin: Okay, so we should start a plan this week, or… and then I think this task will trickle into next week, if not more, because then they will need to approve the costs and all that again.
116 00:09:19.250 ⇒ 00:09:19.810 Amber Lin: Okay.
117 00:09:21.610 ⇒ 00:09:26.119 Amber Lin: Who can own the initial plant?
118 00:09:27.500 ⇒ 00:09:29.129 Amber Lin: A AI draft.
119 00:09:30.670 ⇒ 00:09:33.880 Casie Aviles: That’s gonna be for… This week, right?
120 00:09:33.880 ⇒ 00:09:34.600 Samuel Roberts: Right?
121 00:09:34.600 ⇒ 00:09:35.380 Mustafa Raja: Yeah.
122 00:09:35.670 ⇒ 00:09:36.220 Amber Lin: Yeah.
123 00:09:36.350 ⇒ 00:09:38.700 Amber Lin: So we can get someone to review.
124 00:09:39.630 ⇒ 00:09:41.109 Samuel Roberts: Just put a plan together for the migration.
125 00:09:41.110 ⇒ 00:09:41.620 Casie Aviles: Yeah.
126 00:09:41.620 ⇒ 00:09:42.780 Samuel Roberts: model migration?
127 00:09:43.670 ⇒ 00:09:48.710 Mustafa Raja: I think, here we are just going to ask them to host some models, right? I guess I can take that over.
128 00:09:49.360 ⇒ 00:09:54.079 Samuel Roberts: Yeah, I guess, I guess, yeah, the question there is… yeah, do they just want OpenAI?
129 00:09:55.170 ⇒ 00:10:02.540 Samuel Roberts: We need to probably… As part of that plan, test some new models because of the… discontinuation.
130 00:10:02.540 ⇒ 00:10:06.310 Mustafa Raja: Retiring of the 4, right?
131 00:10:06.310 ⇒ 00:10:08.340 Samuel Roberts: Yeah, so.
132 00:10:09.560 ⇒ 00:10:12.700 Mustafa Raja: I saw that the automated Kimi 2.5, right?
133 00:10:13.310 ⇒ 00:10:15.659 Samuel Roberts: Yeah, there’s a PR for that, I have to get through that.
134 00:10:15.660 ⇒ 00:10:16.410 Mustafa Raja: Yeah, I’m excited.
135 00:10:16.410 ⇒ 00:10:16.780 Samuel Roberts: But yeah.
136 00:10:16.780 ⇒ 00:10:17.839 Mustafa Raja: I’ve heard good things.
137 00:10:17.840 ⇒ 00:10:21.209 Samuel Roberts: Yeah, I’ve been playing with that with, Open Claw. It’s been pretty good.
138 00:10:21.210 ⇒ 00:10:22.010 Mustafa Raja: Oh.
139 00:10:22.470 ⇒ 00:10:29.110 Mustafa Raja: I might set up open claw for… for… for a task that my father does repeatedly for his business.
140 00:10:29.110 ⇒ 00:10:29.760 Samuel Roberts: Yeah?
141 00:10:29.940 ⇒ 00:10:31.619 Mustafa Raja: I might automate that for him.
142 00:10:32.570 ⇒ 00:10:35.070 Samuel Roberts: I have, I have plenty of thoughts if you’re curious, but…
143 00:10:35.070 ⇒ 00:10:38.330 Mustafa Raja: Oh, yeah, so, maybe,
144 00:10:38.330 ⇒ 00:10:41.699 Samuel Roberts: Friday, we’ll meet and talk about that, then… Sounds good. If you have time.
145 00:10:43.520 ⇒ 00:10:49.589 Samuel Roberts: Cool. Okay, yeah, let’s say, now, is it just open, like, is Azure gonna not have 4.0 anymore?
146 00:10:51.060 ⇒ 00:10:58.440 Mustafa Raja: Utam sent something about that, I think Azure isn’t going to have that. Let me, let me, let me search it up.
147 00:10:59.000 ⇒ 00:11:03.420 Mustafa Raja: Let me search it up. Yeah, we retired March 31st, yeah, okay.
148 00:11:04.550 ⇒ 00:11:08.099 Samuel Roberts: Yeah, so, we might even wanna…
149 00:11:08.240 ⇒ 00:11:13.460 Samuel Roberts: I can maybe make a few tickets to just explore that, too, so I don’t know if we want to make a…
150 00:11:13.850 ⇒ 00:11:16.520 Samuel Roberts: The migrate plan is fine for now, we’ll start with that.
151 00:11:17.700 ⇒ 00:11:18.310 Amber Lin: Okay.
152 00:11:18.970 ⇒ 00:11:21.480 Amber Lin: But yeah, there might have to be a few more tests.
153 00:11:21.480 ⇒ 00:11:22.949 Samuel Roberts: Make sure things behave alright.
154 00:11:28.350 ⇒ 00:11:29.870 Amber Lin: Oh.
155 00:11:32.400 ⇒ 00:11:35.600 Mustafa Raja: 1738 is done, if you want to market.
156 00:11:35.600 ⇒ 00:11:38.760 Amber Lin: 17… Wait, where?
157 00:11:38.760 ⇒ 00:11:40.540 Mustafa Raja: Yeah, this one, this bike.
158 00:11:41.430 ⇒ 00:11:42.320 Amber Lin: Ugh.
159 00:11:46.040 ⇒ 00:11:46.680 Amber Lin: Mmm…
160 00:11:57.700 ⇒ 00:11:59.050 Amber Lin: We bet.
161 00:12:00.110 ⇒ 00:12:01.609 Amber Lin: I’m gonna move it here.
162 00:12:02.430 ⇒ 00:12:05.700 Amber Lin: What about this?
163 00:12:06.830 ⇒ 00:12:11.449 Amber Lin: Yeah… Is that part of the AI migration plan?
164 00:12:11.950 ⇒ 00:12:12.810 Samuel Roberts: Oh, yeah.
165 00:12:13.420 ⇒ 00:12:17.630 Mustafa Raja: This should be in AI internal team, to be honest.
166 00:12:17.630 ⇒ 00:12:18.240 Amber Lin: Okay.
167 00:12:18.240 ⇒ 00:12:20.420 Samuel Roberts: That’s true, it’s not just affecting Andy, but yeah.
168 00:12:21.080 ⇒ 00:12:21.930 Mustafa Raja: Yeah.
169 00:12:22.460 ⇒ 00:12:24.709 Mustafa Raja: I’ll move this, I’ll move this over.
170 00:12:25.320 ⇒ 00:12:25.950 Amber Lin: Okay.
171 00:12:28.410 ⇒ 00:12:28.980 Mustafa Raja: Yeah, yeah, yeah.
172 00:12:28.980 ⇒ 00:12:30.369 Amber Lin: Alright, T.
173 00:12:31.600 ⇒ 00:12:38.399 Amber Lin: Move issue. Okay, it’s gone. That we’re not… we’re doing later,
174 00:12:42.550 ⇒ 00:12:43.830 Amber Lin: Okay.
175 00:12:44.470 ⇒ 00:12:46.390 Amber Lin: This, assigned to you.
176 00:12:46.540 ⇒ 00:12:51.980 Amber Lin: That will go later. I don’t honestly feel like it’s too generic.
177 00:12:52.260 ⇒ 00:13:00.510 Amber Lin: Okay, so now I want to look at the central docs stuff, or I guess the transcript. Are you still blocked on that?
178 00:13:01.810 ⇒ 00:13:05.139 Amber Lin: I saw some emails about the transcript stuff.
179 00:13:05.780 ⇒ 00:13:09.279 Samuel Roberts: Yes, I, I’m st… so…
180 00:13:09.460 ⇒ 00:13:15.169 Samuel Roberts: I’m still waiting to hear back from 8x8. I… that was… I sent another one. Tim…
181 00:13:15.790 ⇒ 00:13:20.210 Samuel Roberts: Basically, it’s still kind of trying to figure out the total,
182 00:13:21.180 ⇒ 00:13:26.499 Samuel Roberts: Rate limit, which is only really important for the backfill, which would be the final, like…
183 00:13:27.530 ⇒ 00:13:34.529 Samuel Roberts: everything. So I can still be doing, like, small tests and stuff. When we had a meeting last week, they mentioned something about queues.
184 00:13:34.910 ⇒ 00:13:37.390 Samuel Roberts: In order to filter out,
185 00:13:38.430 ⇒ 00:13:41.539 Samuel Roberts: Out… Outbound calls versus inbound calls.
186 00:13:42.090 ⇒ 00:13:46.239 Samuel Roberts: I don’t… I haven’t heard back about that, I guess, so…
187 00:13:47.810 ⇒ 00:13:50.370 Samuel Roberts: Those are kind of the two things that are holding it up right now.
188 00:13:53.300 ⇒ 00:13:54.769 Samuel Roberts: Okay. So I think the…
189 00:13:54.770 ⇒ 00:13:55.920 Amber Lin: Sounds good.
190 00:13:55.920 ⇒ 00:14:03.400 Samuel Roberts: I… yeah, the 8x8, I’m… just keep following up with him. The other side of it is… which…
191 00:14:04.500 ⇒ 00:14:10.630 Samuel Roberts: I mean, I can do a little bit of exploration and maybe figure out the cues, and maybe come up with a better question to more formally ask.
192 00:14:11.040 ⇒ 00:14:16.849 Samuel Roberts: client, let me… let me do that today, and I’ll get back to you about if I need more information from them.
193 00:14:17.360 ⇒ 00:14:18.150 Amber Lin: Okay.
194 00:14:18.480 ⇒ 00:14:19.610 Amber Lin: Sounds good.
195 00:14:20.790 ⇒ 00:14:30.809 Samuel Roberts: Yeah, I think Janiece had mentioned it, but Yvette wasn’t at the meeting, and then there was an email I saw where she was… wasn’t sure what was going on with that, and yeah, so I’ll do a little more digging, because to be honest, the APIs is not.
196 00:14:31.350 ⇒ 00:14:33.160 Samuel Roberts: the clearest thing.
197 00:14:39.140 ⇒ 00:14:39.610 Samuel Roberts: So, yeah.
198 00:14:39.610 ⇒ 00:14:41.560 Amber Lin: Message,
199 00:15:06.350 ⇒ 00:15:10.770 Amber Lin: Where are we? Oh, we’re here. Okay, so I will…
200 00:15:10.990 ⇒ 00:15:15.469 Amber Lin: I think we can expand that to this full week, so hopefully we get that done.
201 00:15:15.660 ⇒ 00:15:24.740 Amber Lin: On the central dog stuff, I know you guys probably talked about it earlier today, but are we still on track to finish this?
202 00:15:25.250 ⇒ 00:15:28.800 Amber Lin: And what’s the progress there was blocking? Who can help?
203 00:15:29.640 ⇒ 00:15:34.329 Samuel Roberts: Yeah, we… actually, I wanted to talk through the plan for the contradictory and duplicate content detector.
204 00:15:35.320 ⇒ 00:15:40.150 Samuel Roberts: Safa has. It might be good for all of us to go through that a little bit, just to make sure it makes sense.
205 00:15:41.840 ⇒ 00:15:45.419 Amber Lin: Yeah, let’s go through that after we look through these.
206 00:15:45.420 ⇒ 00:15:46.290 Samuel Roberts: Personal, yeah.
207 00:15:46.290 ⇒ 00:15:47.179 Amber Lin: I think…
208 00:15:47.180 ⇒ 00:15:47.979 Samuel Roberts: Yeah, we were gonna talk about.
209 00:15:47.980 ⇒ 00:15:50.740 Amber Lin: I made some progress there, right?
210 00:15:53.240 ⇒ 00:16:02.030 Amber Lin: Sorry, what’s the progress for these? I know Pranav was working on this, and I don’t know if these have been started.
211 00:16:02.030 ⇒ 00:16:03.180 Samuel Roberts: These two.
212 00:16:04.730 ⇒ 00:16:07.160 Samuel Roberts: No, I don’t think so, I think we’re gonna have to move those along.
213 00:16:08.740 ⇒ 00:16:09.770 Samuel Roberts: Okay. Sop.
214 00:16:10.360 ⇒ 00:16:12.910 Samuel Roberts: water… I would.
215 00:16:13.570 ⇒ 00:16:14.060 Amber Lin: Move that.
216 00:16:14.060 ⇒ 00:16:16.820 Samuel Roberts: Yeah, just move that to… move that to next week.
217 00:16:16.820 ⇒ 00:16:17.520 Amber Lin: Okay.
218 00:16:17.690 ⇒ 00:16:19.339 Amber Lin: And then… this…
219 00:16:19.340 ⇒ 00:16:20.010 Samuel Roberts: Together, right?
220 00:16:20.430 ⇒ 00:16:25.579 Samuel Roberts: Any sales sets that run weekly? I don’t think any… no, we haven’t looked at that yet.
221 00:16:27.410 ⇒ 00:16:28.050 Amber Lin: Hmm…
222 00:16:32.460 ⇒ 00:16:37.530 Amber Lin: Okay, so let’s… let’s then talk about… These two.
223 00:16:38.300 ⇒ 00:16:45.290 Amber Lin: Go ahead and share. I need to hop at 10.30, but then the meeting room would be available.
224 00:16:47.720 ⇒ 00:16:50.039 Samuel Roberts: Yeah, Mustafa, do you want to share that.
225 00:16:52.040 ⇒ 00:16:55.830 Mustafa Raja: Yes, I see the screen. Give me a moment.
226 00:16:55.830 ⇒ 00:16:57.220 Samuel Roberts: Sure. Sure.
227 00:16:58.990 ⇒ 00:16:59.859 Mustafa Raja: I don’t know.
228 00:17:12.869 ⇒ 00:17:16.009 Mustafa Raja: Okay, so we have this plan, right? So…
229 00:17:16.140 ⇒ 00:17:35.200 Mustafa Raja: What essentially, this plan is proposing that we do, a combination of some, programmatic parsing through the document, and then, creating some lists or maps, and then passing those maps for, to LLM for, detection of…
230 00:17:35.250 ⇒ 00:17:39.650 Mustafa Raja: Possible, contradictions and duplications.
231 00:17:40.250 ⇒ 00:17:53.940 Mustafa Raja: And that’s pretty much it here. And these are some examples that, these are some examples of duplications throughout this mechanical document that we can consolidate. And we have some…
232 00:17:54.060 ⇒ 00:17:56.369 Mustafa Raja: Contradictions, also, in prices.
233 00:17:57.620 ⇒ 00:17:59.080 Mustafa Raja: In mechanical document.
234 00:18:00.940 ⇒ 00:18:01.770 Mustafa Raja: Yeah.
235 00:18:02.580 ⇒ 00:18:06.119 Samuel Roberts: Okay, so this is, ideally, if we…
236 00:18:06.300 ⇒ 00:18:11.470 Samuel Roberts: Build this to process the mechanical that could process the other ones too, right?
237 00:18:11.710 ⇒ 00:18:20.240 Mustafa Raja: Yeah, yeah, yeah, that’s just building it, and if we just… if we test it and it’s good, this should work with all others, too.
238 00:18:20.780 ⇒ 00:18:22.399 Samuel Roberts: Okay, great.
239 00:18:23.530 ⇒ 00:18:28.989 Samuel Roberts: I looked through it a little bit. I think it… I mean, it makes sense. I guess I am curious about…
240 00:18:31.250 ⇒ 00:18:39.919 Samuel Roberts: What… yeah, what is the… Like, is there a good human-in-the-loop step to be like…
241 00:18:40.440 ⇒ 00:18:43.959 Samuel Roberts: This is duplicate or contradictory, and this is the one that’s important.
242 00:18:45.640 ⇒ 00:18:49.869 Mustafa Raja: I think the output is just going to be a proposed document, right?
243 00:18:50.040 ⇒ 00:18:50.630 Mustafa Raja: It’s not gonna.
244 00:18:50.630 ⇒ 00:18:51.920 Samuel Roberts: Oh, okay, okay.
245 00:18:52.490 ⇒ 00:18:56.570 Samuel Roberts: I will say I skimmed a little bit because it was pretty long, but yes. Okay, that makes sense.
246 00:18:56.800 ⇒ 00:19:01.460 Samuel Roberts: There was something else I had, hold on, let me…
247 00:19:11.320 ⇒ 00:19:12.160 Samuel Roberts: Fine…
248 00:19:20.420 ⇒ 00:19:23.930 Samuel Roberts: No, okay, never mind, I was just looking at what I had typed up, but I think…
249 00:19:24.100 ⇒ 00:19:29.469 Samuel Roberts: Yeah, I mean, how do you feel about… About running this, and…
250 00:19:29.470 ⇒ 00:19:31.910 Mustafa Raja: Yeah, so, my only…
251 00:19:32.510 ⇒ 00:19:39.850 Mustafa Raja: My only speculation would be… so, so the first pass that it’s suggesting is… let me…
252 00:19:43.270 ⇒ 00:19:56.270 Mustafa Raja: So the first pass that it is suggesting is that, we… we generate summaries, of each of the sections, right? And then we pass those summaries to the… what’s it called?
253 00:19:57.500 ⇒ 00:20:15.590 Mustafa Raja: to the LLM, and then LLM decides, okay, which sections do it… does it want to, check for, what’s it called? Duplications and contradictions, right? So, the only concern that I have is, it might… it might… in those summaries, it might miss.
254 00:20:15.950 ⇒ 00:20:32.299 Mustafa Raja: Some of the, you know, context, or some of the duplications or contradictions, that’s… that’s pretty much it. I feel, if there is a way where we could pass the whole document, that’s just going to make our lives easier. But, this is a big, big document, 8,000 lines.
255 00:20:32.300 ⇒ 00:20:38.560 Samuel Roberts: Right. So was it summarizing, where was that?
256 00:20:40.530 ⇒ 00:20:43.190 Samuel Roberts: Normalized area…
257 00:20:46.840 ⇒ 00:20:53.819 Mustafa Raja: I think I did show this concern to the, to Cursor also, and it might have removed that.
258 00:20:54.660 ⇒ 00:21:01.329 Mustafa Raja: Oh, okay. Because I also shared that concern that, you know, some of this isn’t going to work, we need the whole thing.
259 00:21:01.330 ⇒ 00:21:09.559 Samuel Roberts: Yeah, okay, then maybe that’s fine then. I mean, what… I guess the question becomes… Context window, right?
260 00:21:11.180 ⇒ 00:21:12.739 Samuel Roberts: And how do we know?
261 00:21:14.580 ⇒ 00:21:19.270 Samuel Roberts: How many tokens the… the whole mechanical doc in Markdown is?
262 00:21:21.160 ⇒ 00:21:26.749 Mustafa Raja: I don’t know, it’s above, 250 pages, right?
263 00:21:28.010 ⇒ 00:21:30.600 Mustafa Raja: token length…
264 00:21:31.630 ⇒ 00:21:34.230 Samuel Roberts: Yeah, I mean, there might even be a… just a thing you can paste in.
265 00:21:38.960 ⇒ 00:21:41.439 Mustafa Raja: This could be a rough estimate for them.
266 00:21:41.440 ⇒ 00:21:43.630 Samuel Roberts: Alright, so I’ll call it 100,000, that’s fine, okay.
267 00:21:49.760 ⇒ 00:21:51.700 Samuel Roberts: Million token… okay.
268 00:21:58.290 ⇒ 00:22:01.019 Samuel Roberts: Yeah, can you jump back to the plan real quick?
269 00:22:06.280 ⇒ 00:22:07.970 Samuel Roberts: Where’s the docs? Yeah…
270 00:22:15.860 ⇒ 00:22:16.760 Samuel Roberts: Hmm.
271 00:22:17.260 ⇒ 00:22:19.610 Mustafa Raja: Yeah, you see a map of somebody passed.
272 00:22:19.610 ⇒ 00:22:20.400 Samuel Roberts: Yeah…
273 00:22:20.880 ⇒ 00:22:24.930 Mustafa Raja: This is… we need a good strategy.
274 00:22:26.310 ⇒ 00:22:31.320 Mustafa Raja: For AI to know which sections is it going to check, for duplicates.
275 00:22:31.800 ⇒ 00:22:32.610 Samuel Roberts: That’s a good…
276 00:22:33.780 ⇒ 00:22:39.729 Samuel Roberts: thought there, what is the… do we have a sense, and, I mean, you can maybe dig into the… the…
277 00:22:40.320 ⇒ 00:22:42.060 Samuel Roberts: The logs a little bit, but just…
278 00:22:42.530 ⇒ 00:22:48.620 Samuel Roberts: you know, anecdotally, do we have a sense of what types of contradictory data? Like, I know pricing is sometimes a thing.
279 00:22:48.620 ⇒ 00:22:49.340 Mustafa Raja: Yeah.
280 00:22:50.740 ⇒ 00:22:53.259 Samuel Roberts: I know I saw some stuff about the scheduling.
281 00:22:54.650 ⇒ 00:23:00.270 Samuel Roberts: Are there other things we can highlight specifically to make sure are included in the summary that you guys can think of?
282 00:23:03.510 ⇒ 00:23:05.689 Mustafa Raja: I haven’t been up to speed with the triages.
283 00:23:07.050 ⇒ 00:23:07.750 Samuel Roberts: Okay.
284 00:23:09.250 ⇒ 00:23:11.099 Mustafa Raja: But I can take a look after the meeting.
285 00:23:11.100 ⇒ 00:23:13.800 Samuel Roberts: Alright, any thoughts, Casey, Amber, on, like…
286 00:23:14.520 ⇒ 00:23:20.869 Samuel Roberts: Besides, like, pricing and maybe the scheduling, what kind of issues the duplication and contradictory stuff might…
287 00:23:21.200 ⇒ 00:23:24.200 Samuel Roberts: need to be… Like, tuned for?
288 00:23:25.050 ⇒ 00:23:30.090 Amber Lin: I’m also wanting to ask about, like, overall…
289 00:23:30.120 ⇒ 00:23:43.830 Amber Lin: Similarity of certain sections, because, like, some sections are, how to schedule this for this specific service, and 80% of that is the same as scheduling for another service.
290 00:23:43.830 ⇒ 00:23:56.969 Amber Lin: Like, is it better for the AI if we just spell it out in a completely different section, or is it better for AI if we have, if we combine those sections that’s 80% similar, and then have the differences spelled out?
291 00:23:56.970 ⇒ 00:23:57.480 Samuel Roberts: Yeah.
292 00:23:58.220 ⇒ 00:24:06.879 Amber Lin: that’s… that’s more of, like, structural changes. So, like, these are minor, I would say, accuracy issues.
293 00:24:06.880 ⇒ 00:24:07.580 Samuel Roberts: Mmm.
294 00:24:07.580 ⇒ 00:24:13.609 Amber Lin: But then there’s the structural stuff that I… I don’t yet know how to use AI to detect.
295 00:24:14.250 ⇒ 00:24:16.110 Samuel Roberts: Yeah, no, I hear that.
296 00:24:16.110 ⇒ 00:24:18.140 Amber Lin: Similarity scores, or…
297 00:24:18.260 ⇒ 00:24:26.479 Amber Lin: Like, would it tell us, hey, you have… like, on an overall heat map, you have a lot of pricing here, you have a lot of pricing there.
298 00:24:26.820 ⇒ 00:24:30.779 Amber Lin: scattered, you should combine that, like, I don’t… how would we do that?
299 00:24:30.780 ⇒ 00:24:46.099 Mustafa Raja: Yeah, so… so pricing, we do see that the pricing is a lot spread, in the document, right? And it’s suggesting that we should put it in, in the, in the start as this, and then just reference to this. Yeah.
300 00:24:46.100 ⇒ 00:24:52.890 Mustafa Raja: Okay, yeah, that’s smart. This exact section is repeated 3 times in the rock.
301 00:24:53.310 ⇒ 00:25:04.639 Mustafa Raja: And what it’s saying is, here, that we should write it once for level 1. And level 2 and level 3, where it is being repeated, it should just link to that.
302 00:25:07.310 ⇒ 00:25:09.830 Samuel Roberts: Yeah, I think that’s a good start. I mean, I think the…
303 00:25:10.690 ⇒ 00:25:19.180 Samuel Roberts: you know, this task is looking for the duplicate and the contradictory stuff. The, like, kind of formatting might come as we build that SOP.
304 00:25:19.990 ⇒ 00:25:23.849 Samuel Roberts: Kind of formatting tool, but I think we’ll learn a lot from running this.
305 00:25:24.970 ⇒ 00:25:26.510 Samuel Roberts: So I would say…
306 00:25:28.900 ⇒ 00:25:31.600 Mustafa Raja: The only difference from me is…
307 00:25:32.250 ⇒ 00:25:36.769 Mustafa Raja: correctly identifying which sections we need to look into, you know? For…
308 00:25:37.060 ⇒ 00:25:39.330 Mustafa Raja: Duplicates and stuff like that, you know?
309 00:25:40.970 ⇒ 00:25:41.840 Mustafa Raja: Because…
310 00:25:41.840 ⇒ 00:25:42.410 Samuel Roberts: Which…
311 00:25:44.920 ⇒ 00:25:50.120 Mustafa Raja: Yeah, so contradiction candidates. What these candidates are, just maps.
312 00:25:50.240 ⇒ 00:26:04.360 Mustafa Raja: of what, what sections should the LLM check exactly, you know? So, so the exact content is going to be passed to the LLM,
313 00:26:04.360 ⇒ 00:26:12.909 Mustafa Raja: to check these, to check only those sections for contradictions and duplications, right? So we need to identify those correctly.
314 00:26:17.650 ⇒ 00:26:19.209 Samuel Roberts: Pairs of sets, yeah.
315 00:26:21.280 ⇒ 00:26:26.039 Samuel Roberts: Right, so it’s gonna… Different number for the same contacts.
316 00:26:26.040 ⇒ 00:26:32.060 Mustafa Raja: I think that also depends on how we generate the summaries, right? Summaries should be able to reflect
317 00:26:32.460 ⇒ 00:26:39.630 Mustafa Raja: maybe everything that is present, you know? What are we talking about in that particular section, right?
318 00:26:40.080 ⇒ 00:26:40.690 Samuel Roberts: Right, right.
319 00:26:40.690 ⇒ 00:26:44.319 Mustafa Raja: Are there any pricing mentions, or something like that, you know?
320 00:26:44.320 ⇒ 00:26:44.660 Samuel Roberts: Yeah.
321 00:26:44.660 ⇒ 00:26:47.799 Mustafa Raja: As you mentioned, you know, we should definitely…
322 00:26:48.280 ⇒ 00:26:50.489 Mustafa Raja: Check it with other pricing sections.
323 00:26:51.120 ⇒ 00:26:56.300 Samuel Roberts: Yeah, I would say… Yeah, pricing and scheduling, like, we can maybe create a…
324 00:26:56.950 ⇒ 00:27:06.910 Samuel Roberts: a… you know, essentially an array of, like, specific things we want the LLM to focus on, and if it’s not doing its job well enough in the doc… like, we can maybe add to that as we find more.
325 00:27:07.060 ⇒ 00:27:08.240 Mustafa Raja: Yeah, yeah.
326 00:27:08.240 ⇒ 00:27:13.780 Samuel Roberts: to rerun. I think that’s probably, you know, an iterative process here is probably gonna be… Better than not.
327 00:27:13.940 ⇒ 00:27:18.480 Samuel Roberts: Just because, you know, we know the… the kind of…
328 00:27:18.840 ⇒ 00:27:25.300 Samuel Roberts: Clearest things that are contradictory in pricing and… You know, different,
329 00:27:25.540 ⇒ 00:27:34.789 Samuel Roberts: instructions, but there might be some other things that we find once we’ve cleaned it up a little bit that, oh, this is still confusing, but it’s not pricing, it’s something else. And we can probably rerun it then.
330 00:27:35.070 ⇒ 00:27:35.510 Mustafa Raja: Yep.
331 00:27:36.300 ⇒ 00:27:56.270 Amber Lin: Yeah, I also just thought of something, like, some of the stuff in there is more qualitative, like, how do you do this, and, like, this and that, and some of it is actually pretty quantitative. It’s either, like, price numbers, schedule at this time.
332 00:27:56.720 ⇒ 00:27:59.220 Amber Lin: Stuff of it is just concrete.
333 00:27:59.340 ⇒ 00:28:10.529 Amber Lin: answered. Maybe that will be easier to standardize, like, I’ll schedule on this person only, or do this, like, how… is there a way we could…
334 00:28:10.800 ⇒ 00:28:22.230 Amber Lin: separate by that? Would that make it easier in a talk? Like, procedural, maybe procedural versus, like, qual… Definite answers of, this is this.
335 00:28:22.230 ⇒ 00:28:29.340 Samuel Roberts: Yeah, that’s a… that’s a good way to think about it. So maybe we, you know, we look for the specific things that are literal, like…
336 00:28:29.650 ⇒ 00:28:37.509 Samuel Roberts: contradictions, like the pricing or an instruction that’s, like, opposite or something. But then, for broader ones, we might need to…
337 00:28:38.600 ⇒ 00:28:40.730 Samuel Roberts: Well, hmm, sorry, let me think.
338 00:28:45.080 ⇒ 00:28:48.200 Samuel Roberts: Yeah, we might have to do kind of a more general pass as well.
339 00:28:48.360 ⇒ 00:28:52.510 Samuel Roberts: But probably after we remove the kind of clear, quantitative one.
340 00:28:52.510 ⇒ 00:28:52.980 Amber Lin: Yeah.
341 00:28:52.980 ⇒ 00:28:53.710 Samuel Roberts: Let’s see…
342 00:28:54.660 ⇒ 00:29:02.090 Amber Lin: Yeah, so I guess this one right now, this ticket, we were looking at the specifics.
343 00:29:02.170 ⇒ 00:29:20.459 Amber Lin: I think we would eventually categorize or reorganize the document by, like, this section is for procedural stuff, and that section is for, like, concrete answers. I think that would consolidate a lot of stuff after we do that.
344 00:29:21.120 ⇒ 00:29:21.700 Samuel Roberts: Yeah.
345 00:29:22.250 ⇒ 00:29:22.820 Amber Lin: Okay.
346 00:29:22.820 ⇒ 00:29:25.640 Samuel Roberts: Yeah, I think you’re absolutely right, this is kind of step one of it, but…
347 00:29:25.910 ⇒ 00:29:29.450 Amber Lin: Okay, so what would be the next steps here?
348 00:29:30.110 ⇒ 00:29:33.439 Samuel Roberts: I mean, I think if we’re good with this plan,
349 00:29:34.180 ⇒ 00:29:37.290 Samuel Roberts: Mustafa kind of runs with it and sees what kind of,
350 00:29:38.020 ⇒ 00:29:43.500 Samuel Roberts: output comes from, running this with cursor and iterating on that until it…
351 00:29:44.020 ⇒ 00:29:51.470 Samuel Roberts: You know, gives good outputs, basically, for the mechanical dock, and then maybe we try that against the other docks without…
352 00:29:52.110 ⇒ 00:29:54.150 Samuel Roberts: Any kind of prep on them.
353 00:29:57.190 ⇒ 00:29:58.450 Samuel Roberts: Does that sound good?
354 00:29:59.150 ⇒ 00:29:59.800 Mustafa Raja: Yep.
355 00:30:00.350 ⇒ 00:30:00.960 Samuel Roberts: Okay.
356 00:30:04.100 ⇒ 00:30:07.780 Samuel Roberts: Yeah, I would see… I would try to run with this, pass this in.
357 00:30:08.180 ⇒ 00:30:12.850 Samuel Roberts: You know, maybe mention specifically, like, pricing and scheduling and…
358 00:30:13.240 ⇒ 00:30:16.980 Samuel Roberts: Keep those in a way that we can set that, maybe, at the beginning.
359 00:30:17.900 ⇒ 00:30:25.120 Samuel Roberts: I was like… Things to look out for, to, like, loop the agent in, so it’s not just… scanning.
360 00:30:26.790 ⇒ 00:30:31.699 Samuel Roberts: Yeah. I mean, I would say this is a good plan, we just gotta kinda see how it handles it.
361 00:30:32.680 ⇒ 00:30:33.270 Mustafa Raja: Yeah.
362 00:30:36.420 ⇒ 00:30:37.720 Samuel Roberts: Do you feel ready to do that?
363 00:30:38.310 ⇒ 00:30:39.020 Mustafa Raja: Yep.
364 00:30:39.370 ⇒ 00:30:42.130 Samuel Roberts: Okay, that’s the plan then. Sounds good. Yeah.
365 00:30:42.130 ⇒ 00:30:42.480 Amber Lin: Cool.
366 00:30:42.480 ⇒ 00:30:42.880 Mustafa Raja: I’ll do that.
367 00:30:42.880 ⇒ 00:30:47.949 Amber Lin: Do you guys still need to talk about, like, the ticket Panab was working on?
368 00:30:49.800 ⇒ 00:30:54.200 Amber Lin: I know he had some… he needed some help on Snowflake.
369 00:30:54.200 ⇒ 00:30:54.800 Mustafa Raja: Oh, yes.
370 00:30:54.800 ⇒ 00:30:55.440 Samuel Roberts: Just access.
371 00:30:55.440 ⇒ 00:30:56.469 Mustafa Raja: then, man.
372 00:30:56.470 ⇒ 00:30:57.490 Amber Lin: Oh, okay.
373 00:30:57.490 ⇒ 00:31:01.820 Mustafa Raja: Also, I exported the data for last week, and gave it to him.
374 00:31:02.030 ⇒ 00:31:03.080 Mustafa Raja: Okay.
375 00:31:03.080 ⇒ 00:31:03.620 Samuel Roberts: Right.
376 00:31:03.790 ⇒ 00:31:04.880 Mustafa Raja: Should be pretty good.
377 00:31:05.330 ⇒ 00:31:05.830 Samuel Roberts: Okay, cool.
378 00:31:05.830 ⇒ 00:31:06.540 Amber Lin: Sounds good.
379 00:31:07.040 ⇒ 00:31:18.010 Amber Lin: I’ve gotta hop, use this room to discuss if you need, and if we have time, maybe spend, like, 30 minutes, or now we’re looking at triage tickets, that would be nice.
380 00:31:20.060 ⇒ 00:31:20.490 Mustafa Raja: Cook it.
381 00:31:20.490 ⇒ 00:31:21.030 Casie Aviles: You know?
382 00:31:21.340 ⇒ 00:31:22.430 Amber Lin: Alright, thanks guys.
383 00:31:22.630 ⇒ 00:31:23.999 Mustafa Raja: Thank you. See you later, Amber.
384 00:31:26.370 ⇒ 00:31:27.070 Samuel Roberts: Cool.
385 00:31:28.670 ⇒ 00:31:31.589 Samuel Roberts: Alright, yeah, do you think…
386 00:31:32.110 ⇒ 00:31:34.110 Samuel Roberts: You’re good to start running with that, Mustafa.
387 00:31:34.110 ⇒ 00:31:35.010 Mustafa Raja: Yeah.
388 00:31:35.340 ⇒ 00:31:38.689 Samuel Roberts: And then, I guess, I don’t know what’s in triage right now, but,
389 00:31:39.700 ⇒ 00:31:41.680 Samuel Roberts: How bad is that looking right now?
390 00:31:42.750 ⇒ 00:31:43.790 Samuel Roberts: Let me pull it up.
391 00:31:43.790 ⇒ 00:31:44.809 Mustafa Raja: Let me see…
392 00:31:46.150 ⇒ 00:31:47.180 Samuel Roberts: 32…
393 00:31:47.940 ⇒ 00:31:53.900 Mustafa Raja: I have a 32 in the triage, and then they issued more… it should be more in the board.
394 00:31:54.560 ⇒ 00:31:55.780 Samuel Roberts: Oh boy, okay.
395 00:31:57.860 ⇒ 00:32:01.229 Mustafa Raja: I guess, I should be paying more attention to.
396 00:32:03.970 ⇒ 00:32:04.600 Mustafa Raja: To these.
397 00:32:04.600 ⇒ 00:32:05.320 Samuel Roberts: Yeah.
398 00:32:05.660 ⇒ 00:32:06.420 Samuel Roberts: Okay.
399 00:32:07.790 ⇒ 00:32:11.370 Samuel Roberts: Are you guys good to work on that, maybe, for the rest of the working session, then?
400 00:32:11.550 ⇒ 00:32:12.290 Mustafa Raja: Yeah.
401 00:32:13.320 ⇒ 00:32:13.870 Casie Aviles: Hmm.
402 00:32:14.300 ⇒ 00:32:19.399 Casie Aviles: Can… I can take a look at the ones assigned to me.
403 00:32:20.440 ⇒ 00:32:21.120 Samuel Roberts: Okay.
404 00:32:22.950 ⇒ 00:32:30.230 Samuel Roberts: I feel bad that I don’t really have a good sense of, like, how to triage these things, so I don’t know if, like, it’s worth walking me through it so I can help, or if you guys just wanna…
405 00:32:32.130 ⇒ 00:32:37.879 Samuel Roberts: go through it. You know, if I can be a helping hand, but you might have to show me what the kind of normal process is here.
406 00:32:38.130 ⇒ 00:32:39.140 Samuel Roberts: the triage.
407 00:32:40.260 ⇒ 00:32:48.839 Mustafa Raja: Is this a… is this going through that… maybe, sometimes we maybe, take a look at the execution itself.
408 00:32:49.410 ⇒ 00:32:49.800 Samuel Roberts: Okay.
409 00:32:49.800 ⇒ 00:32:57.130 Mustafa Raja: You know, on what it’s doing, and then if it’s giving the wrong answer, why is it doing that? Stuff like that, yeah.
410 00:32:58.100 ⇒ 00:32:58.760 Samuel Roberts: Okay.
411 00:33:00.420 ⇒ 00:33:05.049 Samuel Roberts: Yeah, I mean, I don’t know how long these normally take…
412 00:33:06.860 ⇒ 00:33:09.260 Samuel Roberts: Let me just take a look at a few, make sure I can understand them.
413 00:33:17.090 ⇒ 00:33:24.060 Casie Aviles: I’m gonna share my screen, talk a little bit about… How do I do these?
414 00:33:24.260 ⇒ 00:33:29.000 Casie Aviles: So… I think… Yes, I’m…
415 00:33:29.650 ⇒ 00:33:34.719 Casie Aviles: It would be great, like, if we also have, like… it’s not for now, it’s not…
416 00:33:35.380 ⇒ 00:33:37.550 Casie Aviles: Going to be for now, just an idea, but…
417 00:33:37.980 ⇒ 00:33:42.540 Casie Aviles: I think what helps me understand more what the problem is.
418 00:33:43.040 ⇒ 00:33:45.449 Samuel Roberts: Is if we could, like, auto-label.
419 00:33:46.070 ⇒ 00:33:47.330 Casie Aviles: the tickets…
420 00:33:48.210 ⇒ 00:33:48.900 Samuel Roberts: If they’re.
421 00:33:48.940 ⇒ 00:33:57.509 Casie Aviles: like, a zip code thing, or a text-to-SQL issue, or… Yeah, -oh, let’s see…
422 00:33:57.510 ⇒ 00:33:59.939 Samuel Roberts: Okay, so to do that, we would have to…
423 00:34:00.260 ⇒ 00:34:03.000 Samuel Roberts: Take a look at it, see…
424 00:34:03.540 ⇒ 00:34:09.410 Samuel Roberts: Maybe the feedback, and then maybe be able to actually check the… Database for names?
425 00:34:10.330 ⇒ 00:34:17.159 Casie Aviles: Yeah, what I do is, yeah, I would go here, and then… I would…
426 00:34:18.300 ⇒ 00:34:23.580 Casie Aviles: test it again, like, I would go to… Where is it? Andy…
427 00:34:25.610 ⇒ 00:34:26.210 Samuel Roberts: Okay.
428 00:34:32.780 ⇒ 00:34:35.369 Casie Aviles: then I would, I would ask, like, the question.
429 00:34:35.600 ⇒ 00:34:41.260 Casie Aviles: Basically, as some triage tickets come in, I do it whenever I get the chance, but…
430 00:34:41.500 ⇒ 00:34:42.080 Samuel Roberts: Right.
431 00:34:42.080 ⇒ 00:34:48.619 Casie Aviles: I would just try to run this again and recreate, like, the issue, and then I would check N8N.
432 00:34:52.460 ⇒ 00:34:55.650 Casie Aviles: Yeah, I would check, like, the latest execution, and…
433 00:35:02.150 ⇒ 00:35:06.210 Casie Aviles: Introad services, I think that’s… Technician.
434 00:35:10.950 ⇒ 00:35:13.650 Samuel Roberts: But they said service, is that something we gotta catch better?
435 00:35:15.490 ⇒ 00:35:21.820 Casie Aviles: Yeah, if I remember correctly, like, service is also, like, an alias for technician.
436 00:35:22.460 ⇒ 00:35:23.200 Samuel Roberts: Okay.
437 00:35:24.670 ⇒ 00:35:27.640 Samuel Roberts: But it didn’t… didn’t catch that, I guess.
438 00:35:30.280 ⇒ 00:35:32.360 Casie Aviles: It says it’s not available.
439 00:35:34.180 ⇒ 00:35:41.770 Samuel Roberts: Well, cause… I think it was, it was it didn’t catch the, service was different, right?
440 00:35:43.210 ⇒ 00:35:44.130 Casie Aviles: Oh, yeah.
441 00:35:44.850 ⇒ 00:35:52.290 Casie Aviles: What’s different, and… It looks like Andy checked, like, a different table since earlier.
442 00:35:52.290 ⇒ 00:35:56.749 Samuel Roberts: So we probably need to… Alias… is there… is that alias in there somewhere already, or…
443 00:35:58.410 ⇒ 00:36:03.429 Casie Aviles: Or for handling aliases, I would go to LAMFUSE, and then I would update the prompt.
444 00:36:04.290 ⇒ 00:36:05.510 Samuel Roberts: Oh, I see, okay.
445 00:36:07.290 ⇒ 00:36:15.629 Casie Aviles: So, yeah, like, I did mention that there are, like, two things that affect the answers if it’s related to the Zips DB, so…
446 00:36:15.810 ⇒ 00:36:20.830 Casie Aviles: Either the data is not… present in the database. Right.
447 00:36:21.380 ⇒ 00:36:22.620 Casie Aviles: Or…
448 00:36:23.030 ⇒ 00:36:30.770 Casie Aviles: It’s a prompting issue. That’s why I also, like, updated the response, because I think one reason why they think…
449 00:36:31.200 ⇒ 00:36:37.910 Casie Aviles: Why they might think that it’s missing is because the past message of Andy says that he could.
450 00:36:37.910 ⇒ 00:36:38.350 Samuel Roberts: Yeah.
451 00:36:38.350 ⇒ 00:36:39.200 Casie Aviles: anything.
452 00:36:39.600 ⇒ 00:36:43.340 Casie Aviles: Right. So I updated the message so that…
453 00:36:43.500 ⇒ 00:36:46.729 Casie Aviles: It says this, like, maybe related to how…
454 00:36:47.240 ⇒ 00:36:52.020 Samuel Roberts: Yeah, the question was interpreted, filters, or availability of the data.
455 00:36:52.280 ⇒ 00:36:55.390 Casie Aviles: But then that’s just a band-aid thing.
456 00:36:55.540 ⇒ 00:36:56.230 Samuel Roberts: Yeah.
457 00:36:59.210 ⇒ 00:37:01.910 Casie Aviles: Yeah, it would be… yeah, here… oh, wait.
458 00:37:06.480 ⇒ 00:37:07.869 Casie Aviles: Project was that?
459 00:37:16.460 ⇒ 00:37:19.609 Casie Aviles: Yeah, it’s this… this one here. This is where I update.
460 00:37:19.610 ⇒ 00:37:20.320 Samuel Roberts: Okay.
461 00:37:24.370 ⇒ 00:37:30.070 Casie Aviles: I would just add, like, instructions wherein… let’s see, vector…
462 00:37:35.680 ⇒ 00:37:39.000 Casie Aviles: Yeah, for example, here… Estimator means.
463 00:37:39.000 ⇒ 00:37:43.060 Samuel Roberts: Inspector, so search for the inspector role instead.
464 00:37:43.360 ⇒ 00:37:46.640 Samuel Roberts: That’s… okay. And so we gotta, basically, service, then?
465 00:37:48.300 ⇒ 00:37:51.029 Casie Aviles: Yeah, that’s… that’s how I would do it. Okay.
466 00:37:51.030 ⇒ 00:37:52.079 Samuel Roberts: Yeah, that makes sense.
467 00:37:56.190 ⇒ 00:38:02.129 Casie Aviles: Yeah, that’s pretty much how I go through it, but it would be great, like, if we could…
468 00:38:02.770 ⇒ 00:38:07.440 Casie Aviles: also, like, auto-tag these, so I… we can see, like, how many… Zip.
469 00:38:07.440 ⇒ 00:38:11.330 Samuel Roberts: Related issues are coming in, or if it’s a central dock issue.
470 00:38:12.380 ⇒ 00:38:13.070 Samuel Roberts: Okay.
471 00:38:13.350 ⇒ 00:38:14.250 Samuel Roberts: So…
472 00:38:14.830 ⇒ 00:38:19.129 Samuel Roberts: I mean, first pass could just be if they ask for a zip, whether or not it finds it.
473 00:38:19.390 ⇒ 00:38:24.710 Samuel Roberts: And we can just tag it with the zip one, and then figure out if it’s a text-to-SQL or a zip code DB issue.
474 00:38:25.110 ⇒ 00:38:28.780 Samuel Roberts: And then if not, we could call it a central dock issue, maybe?
475 00:38:30.720 ⇒ 00:38:31.150 Casie Aviles: Mia.
476 00:38:31.150 ⇒ 00:38:35.010 Samuel Roberts: I mean, are there other forms of issues besides the zips and the central dock, though?
477 00:38:38.100 ⇒ 00:38:40.560 Casie Aviles: Hmm.
478 00:38:42.160 ⇒ 00:38:45.780 Casie Aviles: I think those are, like, the main ones I have at the top of my head.
479 00:38:46.260 ⇒ 00:38:47.420 Samuel Roberts: Okay, yeah.
480 00:38:50.690 ⇒ 00:38:52.949 Samuel Roberts: Okay, yeah, maybe, let me take a…
481 00:38:53.740 ⇒ 00:38:56.250 Samuel Roberts: Think about that, what the best way to just, like…
482 00:38:57.510 ⇒ 00:39:00.080 Samuel Roberts: Triage our triage to label them.
483 00:39:03.310 ⇒ 00:39:09.450 Samuel Roberts: Yeah, let me give some thought to that, let me see… I don’t know, we don’t have anything running that just looks at those coming in, do we?
484 00:39:10.930 ⇒ 00:39:13.930 Casie Aviles: I know that we have a workflow that creates these.
485 00:39:14.510 ⇒ 00:39:18.719 Samuel Roberts: Is it in AA? That’s what I was just about to…
486 00:39:19.210 ⇒ 00:39:20.060 Casie Aviles: Yeah.
487 00:39:20.880 ⇒ 00:39:22.180 Casie Aviles: Let me recall.
488 00:39:48.970 ⇒ 00:39:51.909 Casie Aviles: I think it might be this one.
489 00:39:53.930 ⇒ 00:39:57.240 Casie Aviles: Oh yeah, it is this one, since we have the linear nodes here.
490 00:39:57.240 ⇒ 00:39:59.950 Samuel Roberts: Okay, alright, cool, so which one is that? 77…
491 00:40:03.080 ⇒ 00:40:04.949 Samuel Roberts: It’s called… is it catch?
492 00:40:06.710 ⇒ 00:40:07.939 Samuel Roberts: Okay, I got it.
493 00:40:08.080 ⇒ 00:40:08.890 Samuel Roberts: Cool.
494 00:40:12.150 ⇒ 00:40:13.699 Samuel Roberts: Create an issue…
495 00:40:22.190 ⇒ 00:40:26.950 Samuel Roberts: And then we just added in… no, it’s additional field as description, is there a way to do…
496 00:40:28.500 ⇒ 00:40:33.670 Samuel Roberts: assign name or ID or priority. I don’t see a label here on this node.
497 00:40:38.710 ⇒ 00:40:39.530 Samuel Roberts: Okay.
498 00:40:39.670 ⇒ 00:40:40.510 Samuel Roberts: Well…
499 00:40:43.350 ⇒ 00:40:48.270 Casie Aviles: Yeah, we might need to do… well, we have priority, I think, but…
500 00:40:48.800 ⇒ 00:40:52.299 Casie Aviles: We may need to do, like, custom API call.
501 00:40:52.720 ⇒ 00:40:54.929 Samuel Roberts: Yeah, I was gonna say, so where’s this…
502 00:40:55.740 ⇒ 00:41:00.780 Samuel Roberts: This whole flow is just the feedback for… In general?
503 00:41:02.120 ⇒ 00:41:05.549 Casie Aviles: Schedule trigger… what’s this other flow at the bottom here?
504 00:41:06.650 ⇒ 00:41:08.030 Casie Aviles: Oh, this one?
505 00:41:09.090 ⇒ 00:41:12.020 Casie Aviles: I don’t rep… I don’t remember this much.
506 00:41:18.860 ⇒ 00:41:19.610 Samuel Roberts: Okay.
507 00:41:30.440 ⇒ 00:41:32.319 Casie Aviles: Yeah, I mainly worked on this one.
508 00:41:33.080 ⇒ 00:41:33.750 Samuel Roberts: Okay.
509 00:41:33.930 ⇒ 00:41:34.740 Casie Aviles: this.
510 00:41:34.870 ⇒ 00:41:36.690 Casie Aviles: This one above the work.
511 00:41:43.320 ⇒ 00:41:47.439 Casie Aviles: I think I can add maybe an LLM step that will just…
512 00:41:47.440 ⇒ 00:41:47.980 Samuel Roberts: Yeah.
513 00:41:48.160 ⇒ 00:41:49.719 Casie Aviles: categorize them.
514 00:41:54.970 ⇒ 00:41:57.990 Samuel Roberts: It’s just annoying, we can’t just add a label there.
515 00:41:58.730 ⇒ 00:42:02.690 Samuel Roberts: And then, so we’ll need, like, another API call to label that one, maybe?
516 00:42:05.070 ⇒ 00:42:09.840 Casie Aviles: Yeah, I’ll check if there’s, like, a node that can maybe just do, like, update.
517 00:42:10.250 ⇒ 00:42:13.640 Samuel Roberts: Yeah, so I was just looking at the update to see if it had that. I see…
518 00:42:13.950 ⇒ 00:42:16.540 Samuel Roberts: I don’t see labels, that’s so interesting, okay.
519 00:42:35.380 ⇒ 00:42:36.510 Samuel Roberts: Okay.
520 00:42:37.820 ⇒ 00:42:42.669 Samuel Roberts: Well, I mean, at some point we may want to move this out anyway with this migration anyway, right?
521 00:42:43.310 ⇒ 00:42:44.170 Casie Aviles: Yeah.
522 00:42:44.720 ⇒ 00:42:46.520 Samuel Roberts: Maybe we wanna think about that.
523 00:42:47.530 ⇒ 00:42:49.470 Samuel Roberts: That was a way to auto-label.
524 00:42:52.680 ⇒ 00:42:56.610 Casie Aviles: Yeah, I think that should also be a part of our migration, then.
525 00:42:57.390 ⇒ 00:42:58.080 Samuel Roberts: Yeah.
526 00:42:59.140 ⇒ 00:43:02.960 Samuel Roberts: I think we have some stuff for the… Leon…
527 00:43:03.600 ⇒ 00:43:05.720 Samuel Roberts: the feedback in the Gantt, so…
528 00:43:06.130 ⇒ 00:43:10.669 Samuel Roberts: I don’t think we’re getting to it yet, but I’m just trying to figure out if there’s a good way to do this.
529 00:43:25.640 ⇒ 00:43:27.970 Samuel Roberts: Suppose we can… no, priority’s not helpful, never mind.
530 00:43:44.690 ⇒ 00:43:48.960 Casie Aviles: Yeah, I’m pretty sure there’s, like, an API call that can… Update the label.
531 00:43:48.960 ⇒ 00:43:52.919 Samuel Roberts: So, we were doing the same thing, I just didn’t see you on your screen. Yeah.
532 00:43:53.460 ⇒ 00:43:56.690 Casie Aviles: I did the same a while back.
533 00:43:58.460 ⇒ 00:44:00.899 Casie Aviles: Like, I ran a script to do that.
534 00:44:01.490 ⇒ 00:44:02.120 Samuel Roberts: Okay.
535 00:44:08.660 ⇒ 00:44:09.889 Casie Aviles: Let me just…
536 00:44:10.520 ⇒ 00:44:11.860 Samuel Roberts: I wonder if there’s another…
537 00:45:02.860 ⇒ 00:45:05.690 Mustafa Raja: I’m… I’m wondering if… Hmm.
538 00:45:06.080 ⇒ 00:45:07.409 Samuel Roberts: Yeah, I guess…
539 00:45:08.670 ⇒ 00:45:15.980 Mustafa Raja: Yeah, so… yeah, I guess the evals are going to take care of it, but I was just wondering if we could, you know…
540 00:45:17.760 ⇒ 00:45:25.240 Mustafa Raja: Double-check with, with the central docs if the answers that are coming are correct or not, you know.
541 00:45:26.350 ⇒ 00:45:27.439 Mustafa Raja: For the week.
542 00:45:27.700 ⇒ 00:45:31.659 Mustafa Raja: Now, for the week so far that we have… we’ve gotten the queries.
543 00:45:32.040 ⇒ 00:45:37.569 Mustafa Raja: Based, and the ones that are coming from Central Dock. How many of those are correct?
544 00:45:38.060 ⇒ 00:45:39.460 Mustafa Raja: Based on the central doc.
545 00:45:45.070 ⇒ 00:45:47.440 Samuel Roberts: I mean, that’s kind of… Well, I don’t know.
546 00:45:51.340 ⇒ 00:45:54.939 Mustafa Raja: Billy’s thumbs up and thumbs down are really just going to give us too…
547 00:45:54.940 ⇒ 00:45:56.770 Samuel Roberts: Yeah, you’re right, you’re right.
548 00:45:56.770 ⇒ 00:46:00.900 Mustafa Raja: This isn’t going to be a magnified view, right, for everything, right?
549 00:46:01.230 ⇒ 00:46:01.960 Samuel Roberts: Yeah.
550 00:46:34.440 ⇒ 00:46:38.869 Casie Aviles: This is probably not too urgent. I can just, you know, work on…
551 00:46:38.870 ⇒ 00:46:39.440 Samuel Roberts: Okay.
552 00:46:39.560 ⇒ 00:46:42.960 Samuel Roberts: Alright, yeah, I would say focus on the triage, maybe we can make a plan for that to, like…
553 00:46:43.800 ⇒ 00:46:45.099 Casie Aviles: That, that could be…
554 00:46:45.100 ⇒ 00:46:47.420 Samuel Roberts: And we’ll include that, yeah. Okay.
555 00:46:48.550 ⇒ 00:46:49.330 Samuel Roberts: Alright.
556 00:46:49.830 ⇒ 00:46:56.569 Casie Aviles: That’s it, because I still have to test this if I add it now, and I have to put you in the prompt, and…
557 00:46:56.570 ⇒ 00:46:57.550 Samuel Roberts: Don’t worry about it right now.
558 00:46:58.060 ⇒ 00:46:59.439 Samuel Roberts: You’re right, you’re right, okay.
559 00:47:02.740 ⇒ 00:47:04.399 Samuel Roberts: Yeah, we just triage that.
560 00:47:07.000 ⇒ 00:47:08.389 Casie Aviles: No, some deserve.
561 00:47:08.680 ⇒ 00:47:13.710 Casie Aviles: could have been resolved already when I added, so I’ll just… do a quick check.
562 00:47:14.330 ⇒ 00:47:14.980 Samuel Roberts: Okay.
563 00:47:20.160 ⇒ 00:47:23.959 Mustafa Raja: There’s one more thing that I would want. Can I see my screen?
564 00:47:25.090 ⇒ 00:47:26.110 Casie Aviles: Yep, go ahead.
565 00:47:28.450 ⇒ 00:47:33.309 Mustafa Raja: So, so when the charges come in,
566 00:47:33.710 ⇒ 00:47:51.110 Mustafa Raja: these come in as I, as I have, you know, created those, right? So what happens when they get merged into issues is my inbox gets flooded, and it really gets unusable at that point, and I haven’t been able to use it, to be honest. So…
567 00:47:51.110 ⇒ 00:47:51.860 Samuel Roberts: So…
568 00:47:52.330 ⇒ 00:47:58.910 Mustafa Raja: Can we… do we have an account that… or something that nobody’s using, but, you know, for it?
569 00:47:58.910 ⇒ 00:48:01.149 Samuel Roberts: Oh, because you’re the one that… because it’s your account that.
570 00:48:01.150 ⇒ 00:48:04.990 Mustafa Raja: Yeah, it’s my account, it’s my key.
571 00:48:05.850 ⇒ 00:48:08.060 Samuel Roberts: Yeah, is there another… near…
572 00:48:09.620 ⇒ 00:48:10.969 Mustafa Raja: Casey, would you have context on that?
573 00:48:10.970 ⇒ 00:48:11.510 Samuel Roberts: Man.
574 00:48:16.860 ⇒ 00:48:19.720 Samuel Roberts: Let me see if there’s any way to get one that’s not tied to a user.
575 00:48:20.440 ⇒ 00:48:23.629 Casie Aviles: I think the ones I found were all tied to users.
576 00:48:26.140 ⇒ 00:48:27.009 Mustafa Raja: Can we add a guess.
577 00:48:27.010 ⇒ 00:48:28.690 Samuel Roberts: Bot user? Yeah, okay.
578 00:48:31.990 ⇒ 00:48:34.380 Mustafa Raja: I think there is a way.
579 00:48:36.010 ⇒ 00:48:37.600 Casie Aviles: Yeah, let me, let me look at that a little bit.
580 00:48:37.600 ⇒ 00:48:40.979 Samuel Roberts: That actually would be very helpful for some other things, too. We don’t want to keep that…
581 00:48:41.920 ⇒ 00:48:47.600 Mustafa Raja: I think the tickets that get created from the platform is not tied to a user.
582 00:48:48.390 ⇒ 00:48:51.360 Samuel Roberts: Yes, is that polyatomic?
583 00:48:51.670 ⇒ 00:48:57.649 Mustafa Raja: Yes. At that point, it was the only case I used.
584 00:48:59.390 ⇒ 00:49:01.059 Casie Aviles: Yeah, it’d be similar to Discord.
585 00:49:01.060 ⇒ 00:49:03.840 Mustafa Raja: What the point is, is it’s not tied to anyone.
586 00:49:04.370 ⇒ 00:49:08.169 Samuel Roberts: Right, so I’m trying to figure out how is that done right now? Is there a member that is?
587 00:49:09.940 ⇒ 00:49:10.640 Samuel Roberts: Hmm.
588 00:49:12.280 ⇒ 00:49:14.620 Samuel Roberts: Okay, I don’t see a polyatomic member…
589 00:49:16.100 ⇒ 00:49:19.430 Samuel Roberts: But I do see cursor and codecs that are…
590 00:49:30.110 ⇒ 00:49:31.930 Mustafa Raja: Then there’s linear also, right?
591 00:49:32.200 ⇒ 00:49:33.330 Mustafa Raja: Leave it for me.
592 00:49:35.170 ⇒ 00:49:37.080 Samuel Roberts: Yeah, let me…
593 00:49:39.270 ⇒ 00:49:42.559 Samuel Roberts: Alright, let me… let me think about that a little bit today, maybe I can sort that out.
594 00:49:42.760 ⇒ 00:49:43.530 Mustafa Raja: Yeah, yep.
595 00:49:45.020 ⇒ 00:49:49.729 Mustafa Raja: Casey, do you know where this workflow is, in… I believe.
596 00:49:49.730 ⇒ 00:49:50.850 Casie Aviles: Yeah, I just showed it.
597 00:49:51.380 ⇒ 00:49:52.480 Casie Aviles: Oh, okay.
598 00:49:54.090 ⇒ 00:49:54.700 Mustafa Raja: Okay.
599 00:49:55.740 ⇒ 00:50:01.899 Casie Aviles: Yeah, sorry, I used your account there. Before, that was… that was Miguel, the account.
600 00:50:02.320 ⇒ 00:50:06.130 Casie Aviles: And I just switched it to yours, because that’s what I could find.
601 00:50:06.360 ⇒ 00:50:07.989 Mustafa Raja: Oh, yeah, yeah, I understand that.
602 00:50:08.370 ⇒ 00:50:12.149 Samuel Roberts: Well, worst case, we can make a new user, and if not, I’ll figure out another way.
603 00:50:12.150 ⇒ 00:50:16.869 Mustafa Raja: I’m just curious about if we can use the polyatomic one, you know?
604 00:50:17.650 ⇒ 00:50:24.029 Samuel Roberts: Yeah, I would love to know how that works, because if we could use it, then we can make specific ones.
605 00:50:25.370 ⇒ 00:50:28.269 Samuel Roberts: So let me see what I can do here…
606 00:50:30.160 ⇒ 00:50:33.819 Samuel Roberts: Yeah, there’s… Access… past keys, personal keys…
607 00:50:35.500 ⇒ 00:50:42.280 Casie Aviles: there was, Brainforge AI Automations linear, but…
608 00:50:42.600 ⇒ 00:50:46.190 Casie Aviles: This… I probably… I think this was tied to Miguel’s account, so…
609 00:50:46.540 ⇒ 00:50:47.040 Samuel Roberts: Yeah.
610 00:50:47.040 ⇒ 00:50:47.930 Casie Aviles: I’m not sure.
611 00:50:49.500 ⇒ 00:50:50.280 Samuel Roberts: Okay.
612 00:50:52.770 ⇒ 00:50:53.430 Samuel Roberts: Hmm.
613 00:50:55.150 ⇒ 00:50:57.789 Samuel Roberts: Alright, yeah, I’ll do some research on that and see if we can figure that out.
614 00:50:58.290 ⇒ 00:50:58.720 Mustafa Raja: Yeah, yeah, yeah.
615 00:50:58.720 ⇒ 00:51:03.280 Samuel Roberts: into an account, because that’s a really good point. If not, like I said, if you’ll make an account, maybe there’s a linear app.
616 00:51:03.750 ⇒ 00:51:06.669 Samuel Roberts: Or linear apps, like the way we made, you know…
617 00:51:07.060 ⇒ 00:51:09.010 Samuel Roberts: I had to make a GitHub app the other day.
618 00:51:12.550 ⇒ 00:51:16.339 Samuel Roberts: service-type accounts by admins, that’s what I’m looking for, okay.
619 00:51:21.590 ⇒ 00:51:22.260 Samuel Roberts: Okay.
620 00:51:27.210 ⇒ 00:51:31.370 Samuel Roberts: Integrations. Can you build your own integrations, or do they…
621 00:52:14.420 ⇒ 00:52:16.599 Samuel Roberts: Create a new application, you know, there it is.
622 00:52:18.450 ⇒ 00:52:22.940 Samuel Roberts: Okay, let me… let me try this and see if I can make a new application that we could use for everything.
623 00:52:23.470 ⇒ 00:52:24.410 Mustafa Raja: Yep.
624 00:52:27.710 ⇒ 00:52:28.450 Samuel Roberts: Cool.
625 00:52:31.300 ⇒ 00:52:33.340 Mustafa Raja: Where is my browser?
626 00:52:55.050 ⇒ 00:52:57.189 Samuel Roberts: Did not find the page, aw, man.
627 00:53:31.730 ⇒ 00:53:32.820 Samuel Roberts: Oh, you can add…
628 00:53:39.250 ⇒ 00:53:41.010 Samuel Roberts: triage intelligence?
629 00:53:57.250 ⇒ 00:53:58.860 Samuel Roberts: That’s not helpful. Damn it.
630 00:54:14.660 ⇒ 00:54:23.049 Samuel Roberts: Oh, there we go. OAuth flow… to install your agent into a linear workspace in the OAuth flow, add the actor equals app parameter.
631 00:54:24.110 ⇒ 00:54:24.790 Samuel Roberts: Hmm.
632 00:54:26.440 ⇒ 00:54:30.849 Mustafa Raja: What sort of credits does this, does anything take for linear?
633 00:54:38.820 ⇒ 00:54:39.660 Samuel Roberts: We’ll pull that up.
634 00:54:47.380 ⇒ 00:54:48.280 Samuel Roberts: upkeep…
635 00:54:48.560 ⇒ 00:54:50.619 Mustafa Raja: Oh, yeah, AppKey, yeah. API key.
636 00:54:50.620 ⇒ 00:54:52.930 Samuel Roberts: the OAuth 2, though. If you do the OAuth2…
637 00:54:53.330 ⇒ 00:54:57.149 Mustafa Raja: I mean… Hold on, just go back to the OAuth2 real quick.
638 00:54:57.540 ⇒ 00:54:59.570 Samuel Roberts: I just wanna see what options it has here.
639 00:54:59.700 ⇒ 00:55:01.260 Samuel Roberts: Actor, there it is, right there.
640 00:55:01.650 ⇒ 00:55:05.460 Mustafa Raja: Yeah, if you do this application, and then.
641 00:55:05.460 ⇒ 00:55:09.260 Samuel Roberts: So, maybe we try to make an OAuth app for it, and we can do it that way.
642 00:55:09.260 ⇒ 00:55:11.030 Mustafa Raja: Yeah, yeah, that just can keep…
643 00:55:11.030 ⇒ 00:55:12.930 Samuel Roberts: Okay, let me see about getting an OAuth.
644 00:55:19.770 ⇒ 00:55:21.380 Samuel Roberts: Ugh, I’m not logged in here.
645 00:55:25.420 ⇒ 00:55:27.110 Samuel Roberts: Oh, I’m incognito, that’s why.
646 00:55:57.870 ⇒ 00:56:00.639 Samuel Roberts: Client ID, where’d we get that? I don’t understand.
647 00:56:27.860 ⇒ 00:56:29.840 Samuel Roberts: Oh, I bet I don’t have the…
648 00:56:30.360 ⇒ 00:56:33.510 Samuel Roberts: permissions on Linear to go to this page.
649 00:56:33.830 ⇒ 00:56:36.349 Mustafa Raja: Oh, which page is it? I can try going there.
650 00:56:36.350 ⇒ 00:56:38.440 Samuel Roberts: Here, let me, let me send it.
651 00:56:45.910 ⇒ 00:56:47.660 Samuel Roberts: What do you see when you go there?
652 00:56:49.850 ⇒ 00:56:50.680 Mustafa Raja: Nope.
653 00:56:51.430 ⇒ 00:56:57.500 Mustafa Raja: Okay. I can… We might have to ping Rico on this.
654 00:56:58.000 ⇒ 00:56:59.340 Samuel Roberts: Yeah, I was just gonna do that.
655 00:57:00.270 ⇒ 00:57:02.979 Samuel Roberts: Okay, let me take care of that,
656 00:57:05.950 ⇒ 00:57:07.689 Samuel Roberts: And I’ll see what happens there.
657 00:57:08.330 ⇒ 00:57:09.309 Mustafa Raja: Oh, okay.
658 00:57:16.060 ⇒ 00:57:16.980 Mustafa Raja: Okay.
659 00:57:18.000 ⇒ 00:57:19.559 Mustafa Raja: Mmm… Okay.
660 00:57:19.560 ⇒ 00:57:20.480 Samuel Roberts: Very good, though.
661 00:57:20.610 ⇒ 00:57:24.559 Samuel Roberts: You guys… All set, then.
662 00:57:25.350 ⇒ 00:57:26.130 Mustafa Raja: Yep.
663 00:57:26.790 ⇒ 00:57:28.480 Samuel Roberts: Alright, I’ll talk to y’all later.
664 00:57:30.670 ⇒ 00:57:31.400 Samuel Roberts: Bye.
665 00:57:31.400 ⇒ 00:57:32.190 Casie Aviles: Thank you.