Meeting Title: AI Service Standup Date: 2026-02-04 Meeting participants: Casie Aviles, Samuel Roberts, Mustafa Raja, Amber Lin, Gabriel Lam, Uttam Kumaran
WEBVTT
1 00:01:06.560 ⇒ 00:01:07.540 Samuel Roberts: Hey,
2 00:01:11.210 ⇒ 00:01:12.560 Casie Aviles: Oh. Hey, Sam.
3 00:01:13.290 ⇒ 00:01:14.209 Samuel Roberts: How are you?
4 00:01:16.770 ⇒ 00:01:21.190 Casie Aviles: Yeah, doing alright How about you?
5 00:01:22.070 ⇒ 00:01:27.099 Samuel Roberts: Yeah, I’m doing okay, moving a little slow, my… My kid was up.
6 00:01:27.790 ⇒ 00:01:28.650 Casie Aviles: Oh…
7 00:01:29.990 ⇒ 00:01:31.140 Samuel Roberts: A bunch.
8 00:01:32.660 ⇒ 00:01:33.800 Mustafa Raja: Ayy…
9 00:01:37.020 ⇒ 00:01:37.960 Mustafa Raja: Hey.
10 00:01:39.160 ⇒ 00:01:40.170 Casie Aviles: in the sofa.
11 00:01:40.870 ⇒ 00:01:42.280 Mustafa Raja: How’s it going?
12 00:01:44.610 ⇒ 00:01:45.920 Samuel Roberts: It’s going okay.
13 00:01:48.540 ⇒ 00:01:53.980 Samuel Roberts: A little, A little tired today.
14 00:01:56.230 ⇒ 00:02:00.380 Samuel Roberts: But, doing alright.
15 00:02:00.610 ⇒ 00:02:03.480 Samuel Roberts: I’m feeling a little, low.
16 00:02:05.390 ⇒ 00:02:06.250 Samuel Roberts: I was…
17 00:02:07.070 ⇒ 00:02:08.910 Casie Aviles: Yeah, I feel like the week.
18 00:02:08.910 ⇒ 00:02:09.320 Samuel Roberts: Yeah.
19 00:02:09.320 ⇒ 00:02:11.750 Casie Aviles: I mean, as well, it’s similar.
20 00:02:15.810 ⇒ 00:02:16.580 Samuel Roberts: Yeah.
21 00:02:19.920 ⇒ 00:02:20.780 Samuel Roberts: Yeah.
22 00:02:21.330 ⇒ 00:02:22.370 Mustafa Raja: Yeah…
23 00:02:29.580 ⇒ 00:02:30.310 Samuel Roberts: Hey, Amber.
24 00:02:32.770 ⇒ 00:02:33.900 Amber Lin: Hello!
25 00:02:35.650 ⇒ 00:02:36.860 Samuel Roberts: Review this morning?
26 00:02:41.830 ⇒ 00:02:45.000 Amber Lin: Today’s gonna be a very busy day.
27 00:02:45.000 ⇒ 00:02:50.380 Samuel Roberts: Yeah, we were just… I was just saying, I’m… I’m already dragging a little bit, because I was up a lot with my kid last night.
28 00:02:50.720 ⇒ 00:02:51.830 Amber Lin: Oh my…
29 00:02:52.340 ⇒ 00:02:54.420 Samuel Roberts: Yeah, so we’ll see how this goes.
30 00:03:02.970 ⇒ 00:03:04.040 Samuel Roberts: Alright.
31 00:03:11.160 ⇒ 00:03:17.480 Samuel Roberts: I guess let’s maybe start… we can probably do ABC, see if the other people join in.
32 00:03:17.920 ⇒ 00:03:23.380 Samuel Roberts: Yeah, let’s, no.
33 00:03:24.170 ⇒ 00:03:30.349 Samuel Roberts: Sorry, I’m just… my cat’s jumping up and down on my lap, and I’m like, I can’t. Too many, too many things. Okay.
34 00:03:30.670 ⇒ 00:03:31.210 Amber Lin: Yeah.
35 00:03:32.110 ⇒ 00:03:35.550 Amber Lin: what… I mean, I… it’s…
36 00:03:35.550 ⇒ 00:03:35.900 Samuel Roberts: Go ahead.
37 00:03:35.900 ⇒ 00:03:37.130 Amber Lin: Zip code?
38 00:03:37.500 ⇒ 00:03:45.890 Amber Lin: and Central Dock. So, I guess, Casey, how’s the progress on adding the different zip codes?
39 00:03:45.890 ⇒ 00:03:46.410 Casie Aviles: Nope.
40 00:03:47.540 ⇒ 00:03:49.049 Casie Aviles: Hello? Did I get caught up?
41 00:03:49.370 ⇒ 00:03:49.930 Samuel Roberts: You’re…
42 00:03:50.390 ⇒ 00:03:51.790 Amber Lin: Yeah, I can hear you.
43 00:03:51.790 ⇒ 00:03:57.349 Casie Aviles: Oh, yeah, so for the zip code, so I’m… for the mechanical ones, I just…
44 00:03:57.460 ⇒ 00:03:59.550 Casie Aviles: Made updates to the…
45 00:04:00.990 ⇒ 00:04:05.889 Casie Aviles: To the table now, so now it should accommodate for, like, the mechanical dispatch.
46 00:04:08.210 ⇒ 00:04:10.270 Casie Aviles: So, yeah, I’m just going to…
47 00:04:10.860 ⇒ 00:04:19.159 Casie Aviles: make sure that everything’s in there, so I’m… I’m already… I’ve already used… uploaded, like, the mechanical…
48 00:04:20.060 ⇒ 00:04:24.150 Casie Aviles: dispatched to, like, a temporary table, but I have to, like.
49 00:04:24.810 ⇒ 00:04:29.340 Casie Aviles: get it into the main one, so once… so I just had to make sure.
50 00:04:29.910 ⇒ 00:04:36.419 Casie Aviles: That’s kind of our… And then after that, I can start adding, like, the other departments.
51 00:04:37.200 ⇒ 00:04:37.860 Casie Aviles: Okay.
52 00:04:38.200 ⇒ 00:04:54.680 Amber Lin: Sounds good. So after that, I think we’re going to look at lawn to see, if the lawn technicians are in there, because I remember Sam checked and that we’re missing about 60 or 70% of them. So let’s do that after mechanical.
53 00:04:55.760 ⇒ 00:04:56.350 Casie Aviles: Okay.
54 00:04:57.000 ⇒ 00:04:58.370 Amber Lin: Cool.
55 00:04:59.290 ⇒ 00:05:02.439 Casie Aviles: Yeah, I’ll just ping you guys on the channel once I have.
56 00:05:02.440 ⇒ 00:05:20.940 Amber Lin: Yeah, sounds good, because I would like to let the client know, because, like, I think they’re pretty stressed, or, like, they’re pretty concerned, because they’ve been talking about it for a while. On the central doc side, I know Utam also asked in the channel, is there any progress there?
57 00:05:27.010 ⇒ 00:05:28.280 Samuel Roberts: Fahrenheit.
58 00:05:28.510 ⇒ 00:05:29.370 Samuel Roberts: the…
59 00:05:29.370 ⇒ 00:05:30.209 Amber Lin: Yeah, with the
60 00:05:30.690 ⇒ 00:05:40.619 Amber Lin: Yeah, with the automation stuff. I mean, I think there’s no progress, but so I was thinking, like, we can use the working session today.
61 00:05:40.620 ⇒ 00:05:42.520 Samuel Roberts: That’s exactly what I was gonna say, yes.
62 00:05:42.520 ⇒ 00:05:48.350 Amber Lin: Okay, awesome. So, who should be in that working session?
63 00:05:49.950 ⇒ 00:05:55.599 Amber Lin: Let me invite, let me invite Pranav as well.
64 00:05:55.600 ⇒ 00:05:56.730 Samuel Roberts: Yeah, that’s what I was at.
65 00:05:57.270 ⇒ 00:06:03.519 Amber Lin: Enough, and then I don’t think Uoutan will be able to make it, but I’ll say option or forehead.
66 00:06:03.520 ⇒ 00:06:05.079 Samuel Roberts: with something. Yeah, exactly.
67 00:06:05.080 ⇒ 00:06:08.540 Amber Lin: Okay, would you need Casey and Mustafa to be there?
68 00:06:12.810 ⇒ 00:06:14.710 Mustafa Raja: I think I should use it a little more.
69 00:06:14.710 ⇒ 00:06:16.600 Samuel Roberts: Yeah, I think… yeah.
70 00:06:17.740 ⇒ 00:06:19.510 Mustafa Raja: Just whichever one you feel is more comfortable.
71 00:06:19.560 ⇒ 00:06:20.500 Samuel Roberts: Yeah.
72 00:06:20.500 ⇒ 00:06:25.640 Casie Aviles: But, yeah, if you need any context, then I can help, like, provide…
73 00:06:26.560 ⇒ 00:06:30.099 Casie Aviles: Just stay there as I work on the zips as well.
74 00:06:30.900 ⇒ 00:06:46.650 Amber Lin: Cool, okay. Sounds good. I probably can’t be there, because I will be cramming the slides for the final presentation, but, like, I can, I can write the goals, in the Slack channel.
75 00:06:46.930 ⇒ 00:06:48.420 Mustafa Raja: Yeah, that’d be good.
76 00:06:49.200 ⇒ 00:06:50.330 Amber Lin: Cool.
77 00:06:50.330 ⇒ 00:06:54.080 Samuel Roberts: You can look at that document, Mustafa, as well. Awesome.
78 00:06:54.360 ⇒ 00:06:58.549 Amber Lin: Okay, any, like, migration updates, then?
79 00:06:59.310 ⇒ 00:07:15.629 Mustafa Raja: Yes, so yesterday I looked into the… what’s it called? The entry point. It was pointing to, the agent, so I updated it to point it to, the workflow, and then I also worked on the latency threshold.
80 00:07:15.630 ⇒ 00:07:16.330 Samuel Roberts: Yes.
81 00:07:16.330 ⇒ 00:07:24.860 Mustafa Raja: Endpoint, I created a new endpoint, a custom endpoint in Mastra, and I have also linked the Slack bot.
82 00:07:25.040 ⇒ 00:07:30.040 Mustafa Raja: So now, if it, if, any, any…
83 00:07:31.160 ⇒ 00:07:39.749 Mustafa Raja: Execution takes, more than 5 seconds, it’s just going to update this in the, Client ABC Logs channel.
84 00:07:40.010 ⇒ 00:07:49.759 Mustafa Raja: I did a test yesterday, we can look at the format, let me know if this format is good enough, or if you want to make any changes in it, I can do that.
85 00:07:49.760 ⇒ 00:07:51.340 Samuel Roberts: Yeah, I was looking at it, I think it looks…
86 00:07:51.890 ⇒ 00:07:54.969 Samuel Roberts: Very good. I’m trying to think if there’s any other information here we would want.
87 00:07:58.000 ⇒ 00:07:59.660 Samuel Roberts: this session ID…
88 00:08:00.410 ⇒ 00:08:01.220 Mustafa Raja: Is that…
89 00:08:01.220 ⇒ 00:08:07.719 Samuel Roberts: No, no, no, so, yeah, yeah. So, for this session ID, I just updated to the, what’s it called, workflow.
90 00:08:07.970 ⇒ 00:08:10.620 Mustafa Raja: workflow runner ID, you know? So we can…
91 00:08:10.620 ⇒ 00:08:11.060 Samuel Roberts: Oh, okay.
92 00:08:11.420 ⇒ 00:08:22.450 Mustafa Raja: actually extracted. I don’t think this is accurate, this, this isn’t good enough, yeah. So, I’ll update this to be run ID rather than this, this ID.
93 00:08:22.450 ⇒ 00:08:25.559 Samuel Roberts: Yeah, as long as we can just get to those logs in the database.
94 00:08:25.560 ⇒ 00:08:33.779 Mustafa Raja: So the idea is, yeah, so the idea is, we are really replacing, NAT and executions with this stuff, right?
95 00:08:35.130 ⇒ 00:08:41.250 Mustafa Raja: We can directly get to every single log we can get out of it. Yeah.
96 00:08:42.669 ⇒ 00:08:43.209 Samuel Roberts: Okay.
97 00:08:43.460 ⇒ 00:08:52.189 Mustafa Raja: Yeah, so that’s pretty much it for my yesterday. Apart from that, I also worked on the, running dock for GCP services and cost.
98 00:08:52.780 ⇒ 00:08:54.279 Mustafa Raja: I’ll share it soon.
99 00:08:54.280 ⇒ 00:08:54.930 Samuel Roberts: Okay.
100 00:08:57.200 ⇒ 00:08:57.960 Mustafa Raja: offices.
101 00:08:59.200 ⇒ 00:09:01.630 Samuel Roberts: Sorry, I got disconnected for a second, I missed the end of that.
102 00:09:02.230 ⇒ 00:09:04.890 Mustafa Raja: Yeah, I said that I’ll share it, after this meeting.
103 00:09:04.890 ⇒ 00:09:07.569 Samuel Roberts: Okay, cool. Perfect. Alright, yeah, I think.
104 00:09:07.570 ⇒ 00:09:08.860 Mustafa Raja: We can feed them.
105 00:09:10.180 ⇒ 00:09:15.040 Samuel Roberts: Yeah, I… Tim is supposed to get back to me about some other stuff, too, so I haven’t heard back yet, but…
106 00:09:15.040 ⇒ 00:09:21.229 Mustafa Raja: Yeah, I think, email might have been… might be the best, you know, way to communicate. Did they really?
107 00:09:21.810 ⇒ 00:09:22.940 Mustafa Raja: negative.
108 00:09:23.170 ⇒ 00:09:26.439 Samuel Roberts: He had said Slack was better when I talked to him in September.
109 00:09:27.550 ⇒ 00:09:33.349 Mustafa Raja: Okay, okay. Yeah, for recently, we’ve just been communicating on emails.
110 00:09:33.350 ⇒ 00:09:34.570 Samuel Roberts: Okay.
111 00:09:34.570 ⇒ 00:09:43.909 Mustafa Raja: There’s one more thread where he hasn’t come back to us. So, on emails, he just, you know, he replies on emails.
112 00:09:43.910 ⇒ 00:09:45.470 Samuel Roberts: Okay, that’s true, yeah, I…
113 00:09:45.730 ⇒ 00:09:51.269 Samuel Roberts: I was wondering, because it didn’t look like he was even online, so… Okay, I’ll send an email then anyway, because I gotta do that. Okay.
114 00:09:52.470 ⇒ 00:09:53.530 Samuel Roberts: Thank you for that.
115 00:09:54.060 ⇒ 00:09:56.730 Samuel Roberts: Yeah, let me know if industry…
116 00:09:58.330 ⇒ 00:10:05.470 Mustafa Raja: Yeah, let me know if in the same email we want to, you know, ask for this, Cloud SQL stuff.
117 00:10:07.780 ⇒ 00:10:15.479 Samuel Roberts: Yeah… Yeah, once you post that, we’ll put together a message, I’ll include what I’ve got, and then…
118 00:10:16.040 ⇒ 00:10:18.059 Samuel Roberts: We can ask that, too, all in one.
119 00:10:18.600 ⇒ 00:10:19.910 Mustafa Raja: Yeah, okay.
120 00:10:22.650 ⇒ 00:10:29.189 Samuel Roberts: Alright, anything else ABC-wise? We’ll have that working session later, that should be good.
121 00:10:29.300 ⇒ 00:10:31.349 Mustafa Raja: That’s pretty much it from my side, yeah.
122 00:10:32.200 ⇒ 00:10:32.780 Samuel Roberts: Great.
123 00:10:35.500 ⇒ 00:10:36.910 Samuel Roberts: I think that’s good, then.
124 00:10:39.330 ⇒ 00:10:47.180 Samuel Roberts: Alright, I guess let’s… Jump… 2… Well…
125 00:10:47.360 ⇒ 00:10:51.190 Samuel Roberts: I don’t know what to jump to, because we’re missing people that are muted for those ones, but we can talk…
126 00:10:51.330 ⇒ 00:10:54.720 Samuel Roberts: Lilo, quickly.
127 00:10:55.760 ⇒ 00:11:04.059 Samuel Roberts: So the… their prod is good, it seems. They… they’re using it. We got… you got those keys updated, Casey, for Gemini, which I think was kind of the last…
128 00:11:04.430 ⇒ 00:11:05.270 Samuel Roberts: Thing?
129 00:11:05.460 ⇒ 00:11:08.120 Casie Aviles: And yeah, this lock one, I think, as well.
130 00:11:09.300 ⇒ 00:11:16.090 Samuel Roberts: Oh, I’m the Slack one, that’s the one I forgot about, yes, thank you. So yeah, it should be pretty functional here, hopefully get some feedback from them.
131 00:11:17.330 ⇒ 00:11:20.079 Samuel Roberts: As their team starts using it. And I think…
132 00:11:20.560 ⇒ 00:11:24.690 Samuel Roberts: Pranav sorted out some of the data stuff with Shopify.
133 00:11:26.230 ⇒ 00:11:34.150 Casie Aviles: Yes, yeah, we got, we got… yeah, sorry, I was, I was just mentioning that.
134 00:11:34.150 ⇒ 00:11:38.189 Samuel Roberts: Okay, yeah, yeah, yeah, I think it’s just… the data’s matching now, which is good. I don’t know…
135 00:11:38.530 ⇒ 00:11:42.109 Samuel Roberts: If you did anything else after the end of my day yesterday, but…
136 00:11:45.840 ⇒ 00:11:49.149 Samuel Roberts: Okay, I’ll have to connect with him later.
137 00:11:49.400 ⇒ 00:11:52.969 Casie Aviles: Yeah, I don’t think… I also want to sync with…
138 00:11:53.290 ⇒ 00:12:05.640 Casie Aviles: to know, like, where we end up, where we ended up with the data, since we’ve set up, like, Polyatomic already, the free account, and then we were able to successfully sync another one, so…
139 00:12:06.330 ⇒ 00:12:11.539 Samuel Roberts: Okay, so Polytomic is working for Shopify for Newton.
140 00:12:12.380 ⇒ 00:12:19.560 Casie Aviles: Yeah, I think he was going to, match, or, like, update the query to make sure it matches with.
141 00:12:19.560 ⇒ 00:12:23.880 Samuel Roberts: Yeah, yeah, I think we got it working with the airbite data last night, and I was helping him. Okay.
142 00:12:23.880 ⇒ 00:12:24.560 Casie Aviles: Okay.
143 00:12:25.500 ⇒ 00:12:27.210 Samuel Roberts: Yeah, we’ll talk with him later then.
144 00:12:28.890 ⇒ 00:12:31.470 Samuel Roberts: Good.
145 00:12:32.060 ⇒ 00:12:36.510 Samuel Roberts: I guess, we can jump to internal stuff. We’ve got Kevin Utam here.
146 00:12:36.910 ⇒ 00:12:42.110 Samuel Roberts: Yeah, so… I was gonna reply to you earlier on Slack, I saw your, message about the.
147 00:12:42.110 ⇒ 00:12:48.080 Uttam Kumaran: Yeah, I just have to start pushing stuff out, so I think what I’m gonna ask for is, like.
148 00:12:49.140 ⇒ 00:12:51.230 Uttam Kumaran: Just, like…
149 00:12:51.340 ⇒ 00:13:07.189 Uttam Kumaran: keep up. But also, like, it’s gonna be a battle between us, so I feel okay about it. Like, I think it’s a good tension, actually, to have, because I’m learning a lot, and I just really don’t have
150 00:13:07.370 ⇒ 00:13:17.319 Uttam Kumaran: a ton of time, so the time I get to do a lot of this is usually after hours, where I’m like, okay, let me spend, like, an hour and two and pick a team and see how I can solve a few problems.
151 00:13:17.800 ⇒ 00:13:21.710 Uttam Kumaran: So I did get a few things out. I also did push some…
152 00:13:21.710 ⇒ 00:13:22.200 Samuel Roberts: Fuck.
153 00:13:22.200 ⇒ 00:13:33.229 Uttam Kumaran: I feel like innovations to the process. So, one is I enabled BugBot Autofix. So, what it does is when BugBot release, puts in a change request.
154 00:13:33.230 ⇒ 00:13:34.919 Samuel Roberts: Oh, nice, okay.
155 00:13:34.920 ⇒ 00:13:42.120 Uttam Kumaran: it will… it will go ahead and actually fix it in another PR that you can merge into that PR branch.
156 00:13:42.500 ⇒ 00:13:50.159 Uttam Kumaran: So, one thing that’s really tough for me is, like, I’m managing 5 or 6
157 00:13:50.310 ⇒ 00:14:08.370 Uttam Kumaran: feature pushes, like, at a time, and it’s been… it’s really, really annoying to, like, wait for BugBot, and then have to go resolve the fixes. And so, roughly, like, a couple of things that I shared, you know, here in Slack that… that I think is worth all of us
158 00:14:08.520 ⇒ 00:14:13.200 Uttam Kumaran: You know, thinking about, I don’t know if this one, is this the right slot?
159 00:14:13.200 ⇒ 00:14:14.880 Samuel Roberts: Yeah, I saw the labels and stuff.
160 00:14:17.180 ⇒ 00:14:20.349 Uttam Kumaran: Okay, if you can see this. Yeah, so, one is, like.
161 00:14:21.090 ⇒ 00:14:28.680 Uttam Kumaran: I… I… right now, Codex and Cursor and my own CLI were not able to test deploys.
162 00:14:28.780 ⇒ 00:14:37.010 Uttam Kumaran: Before issuing PRs. So, that would… like, I’m thinking about, like, okay, what are the 5 things that would cut out
163 00:14:37.260 ⇒ 00:14:42.250 Uttam Kumaran: Like, paying for any non-technical user is, like, deploy, fix issues.
164 00:14:42.520 ⇒ 00:14:47.169 Uttam Kumaran: So… I do feel like switching to railway is gonna solve this.
165 00:14:47.940 ⇒ 00:14:53.450 Uttam Kumaran: I don’t know whether switching to railway is something that, like, I can literally…
166 00:14:54.390 ⇒ 00:14:57.269 Uttam Kumaran: just tell Codex to go do, or if that’s a more…
167 00:14:57.420 ⇒ 00:14:59.270 Samuel Roberts: Like, severe process.
168 00:14:59.360 ⇒ 00:15:02.249 Uttam Kumaran: ideally, we can do that, and then I feel like
169 00:15:02.850 ⇒ 00:15:09.689 Uttam Kumaran: that we can create a PR commit step that just… that does the deploy locally to check, basically.
170 00:15:10.610 ⇒ 00:15:18.490 Samuel Roberts: Yeah, I think we’re probably good to flip it, if I’m… I mean, it’s one Next app, so it’s not like there’s a bunch of pieces of it.
171 00:15:19.460 ⇒ 00:15:33.680 Samuel Roberts: So we could probably… and all the… as long as the environment variables and everything are set to the Supabase stuff, and all the… all of those are set, it’s not like there’s a database like we have with Lilo, where we gotta manage different… different databases, we really just have one. So we might be able to just spin that up on Railway. I think, really, we gotta figure out…
172 00:15:34.040 ⇒ 00:15:38.209 Samuel Roberts: I mean, I want… is the issue with Codex not able to…
173 00:15:38.750 ⇒ 00:15:39.480 Uttam Kumaran: I don’t know.
174 00:15:39.480 ⇒ 00:15:42.609 Samuel Roberts: build it at cloud on it. Like, I think we just have to get that environment tuned up.
175 00:15:42.610 ⇒ 00:15:45.440 Uttam Kumaran: Yeah. That’s where, like, I noticed that even just…
176 00:15:45.560 ⇒ 00:15:47.450 Uttam Kumaran: Yeah. I’m too dumb to figure this out.
177 00:15:49.090 ⇒ 00:15:51.719 Samuel Roberts: No, no, I’ll take a look at the codex. Oh, go ahead.
178 00:15:51.720 ⇒ 00:15:52.330 Mustafa Raja: I bet.
179 00:15:52.330 ⇒ 00:16:16.090 Mustafa Raja: So, when I ask you to test, what it really does is it only, you know, checks if the linting is correct or not. So, maybe if we explicitly ask it to build and see if build is, you know, working or not, that might solve it, and I’ve seen it once or twice, taking a screenshot. So, to me, it looks like it is able to, you know.
180 00:16:16.090 ⇒ 00:16:17.610 Mustafa Raja: Run it.
181 00:16:17.610 ⇒ 00:16:18.590 Samuel Roberts: I should be.
182 00:16:18.590 ⇒ 00:16:28.399 Mustafa Raja: I think it’s just getting that… I think what we need to do is, get our… get a template of script that we need to give it, so it can work, you know?
183 00:16:29.000 ⇒ 00:16:29.620 Uttam Kumaran: Okay.
184 00:16:30.200 ⇒ 00:16:33.479 Samuel Roberts: No, I think I… I think I know how to do that.
185 00:16:33.740 ⇒ 00:16:39.089 Samuel Roberts: I was looking in there yesterday, because I accidentally kicked off a job to the cloud, and that’s the only environment we have set up in the cloud, and it wasn’t.
186 00:16:39.090 ⇒ 00:16:40.040 Uttam Kumaran: Yes.
187 00:16:40.270 ⇒ 00:16:45.690 Samuel Roberts: So I was like, oh, wait a minute. I think I understand the environments a little bit better now after using the app.
188 00:16:45.850 ⇒ 00:16:47.300 Samuel Roberts: So, I’m gonna work on that.
189 00:16:47.580 ⇒ 00:16:50.469 Mustafa Raja: So you’re working with the Codex app? The desktop one?
190 00:16:50.470 ⇒ 00:16:53.239 Samuel Roberts: I downloaded the… Yeah, I downloaded the new one.
191 00:16:53.800 ⇒ 00:16:55.129 Mustafa Raja: It’s interesting.
192 00:16:55.860 ⇒ 00:17:00.380 Samuel Roberts: Yeah, give it a try. It’s different. It’s, like, it’s way more just agent-focused.
193 00:17:00.690 ⇒ 00:17:03.919 Samuel Roberts: And diffs, and then creating pull requests and commits.
194 00:17:05.099 ⇒ 00:17:09.469 Samuel Roberts: it’s kind of like that agent mode in cursor, but there’s, like, no IDE at all, really.
195 00:17:09.930 ⇒ 00:17:13.779 Samuel Roberts: Okay. And you can just kick off multiple jobs and multiple projects, kind of nice.
196 00:17:14.430 ⇒ 00:17:16.700 Mustafa Raja: Yeah, sounds interesting, I’ll give it a try.
197 00:17:17.450 ⇒ 00:17:25.359 Samuel Roberts: Yeah, and you can run it locally, and you can do work trees, which is the problem I’ve been having with Cursor, and I know what Tom’s been having too, where you can’t, like, run multiple things sometimes.
198 00:17:25.369 ⇒ 00:17:27.469 Mustafa Raja: And it can also run the clouds, so…
199 00:17:27.719 ⇒ 00:17:35.919 Mustafa Raja: Yeah, I have a question for WorkTrees. So with WorkTrees, are we able to work in multiple branches, or it’s like multiple agents on the same branch?
200 00:17:37.290 ⇒ 00:17:43.250 Samuel Roberts: It should be able to be multiple branches, because basically it creates a copy, like a Git copy somewhere else on your file system.
201 00:17:43.390 ⇒ 00:17:45.850 Samuel Roberts: Does work there, and then, like, brings it back.
202 00:17:46.090 ⇒ 00:17:47.349 Mustafa Raja: Is what I understand. Okay.
203 00:17:47.820 ⇒ 00:17:54.560 Samuel Roberts: Okay. So, the problem is that it has to be, like, one Git repo that you’re working in in Cursor. If you have multiple Git repos open, it won’t work.
204 00:17:54.960 ⇒ 00:17:59.459 Mustafa Raja: Okay, okay. Okay. That’s one reason we want to move to that monorepo.
205 00:17:59.790 ⇒ 00:18:09.409 Samuel Roberts: Yeah, so I want to sort that out a little bit anyway, and I think the monorepo will help that too, but yeah, I think getting the codex environment set up will help a lot with that, because then it can do more testing, and…
206 00:18:09.590 ⇒ 00:18:14.369 Samuel Roberts: And just, you know, more stuff automated, even in the cloud, let alone… Yeah.
207 00:18:15.110 ⇒ 00:18:15.700 Gabriel Lam: Yeah.
208 00:18:15.700 ⇒ 00:18:16.910 Samuel Roberts: Yeah, okay, what else?
209 00:18:17.100 ⇒ 00:18:23.880 Gabriel Lam: to, sort of… have a migration sort of plan for the monorepo, because I’m.
210 00:18:23.880 ⇒ 00:18:27.100 Uttam Kumaran: We just need… we just need to push it all into one. There’s… there’s literally, like.
211 00:18:27.100 ⇒ 00:18:34.999 Samuel Roberts: Yeah, I think, I think, I mean, I was gonna ask that, actually, about, do we care about the Git history of the Vault or the platform, or the playbook anymore?
212 00:18:35.000 ⇒ 00:18:35.890 Gabriel Lam: Really?
213 00:18:36.090 ⇒ 00:18:38.350 Uttam Kumaran: Does it… does it… does it matter?
214 00:18:39.230 ⇒ 00:18:41.269 Samuel Roberts: I mean, there’s ways to do, like…
215 00:18:41.950 ⇒ 00:18:42.860 Uttam Kumaran: Bring in the history.
216 00:18:42.860 ⇒ 00:18:50.399 Samuel Roberts: submodules, and, like, maybe we can merge the history in, but I was like, if I’m just gonna move all the files and just push it into the repo and then re… I want to make sure that we move…
217 00:18:50.690 ⇒ 00:18:54.860 Samuel Roberts: the application files with Git, not just with, like.
218 00:18:55.440 ⇒ 00:18:59.139 Uttam Kumaran: Well, what I would suggest is to actually just,
219 00:18:59.640 ⇒ 00:19:03.209 Uttam Kumaran: Move it, and then we can archive the other repos.
220 00:19:03.450 ⇒ 00:19:10.470 Samuel Roberts: Yeah, yeah, I just… I want to make sure that we… when we… because we gotta move the platform logic and, like, other apps into an app folder.
221 00:19:11.040 ⇒ 00:19:16.470 Samuel Roberts: And if we just move those, we’ll lose all the history, because it’ll basically look like files got deleted and created.
222 00:19:16.740 ⇒ 00:19:24.990 Samuel Roberts: And so there’s a specific, like, git move that, like, keeps all the history for the code. The other ones, I just didn’t know if we cared about, so I’m just gonna move those manually, so…
223 00:19:25.410 ⇒ 00:19:34.070 Uttam Kumaran: Well, yeah, if we could just do, like, yeah, if you could just basically copy those in… into platform, and then move…
224 00:19:34.670 ⇒ 00:19:42.000 Uttam Kumaran: Platform app into slash app using the It’s a smart way.
225 00:19:42.310 ⇒ 00:19:43.380 Samuel Roberts: Yeah.
226 00:19:44.610 ⇒ 00:19:45.840 Uttam Kumaran: And not delete.
227 00:19:46.370 ⇒ 00:19:49.990 Samuel Roberts: Okay, great. And can we… if we could do this on Friday?
228 00:19:50.240 ⇒ 00:19:53.159 Uttam Kumaran: Like, after hours, or, like, after 3?
229 00:19:53.650 ⇒ 00:19:58.940 Uttam Kumaran: Because then on Monday, I’ll basically be like, everybody, just pull this one repo, you no longer have to deal with.
230 00:19:58.940 ⇒ 00:19:59.860 Samuel Roberts: Sure, yeah, yeah.
231 00:20:00.140 ⇒ 00:20:07.300 Samuel Roberts: Yeah, I might… I might do it locally to make sure that it works the way I think it does, but then I’ll wait and redo it Friday, if that works.
232 00:20:07.650 ⇒ 00:20:10.480 Uttam Kumaran: Okay. So I’ll try the Heroku to railway.
233 00:20:10.980 ⇒ 00:20:14.730 Uttam Kumaran: if Codex, like, I’ll just be like Sam Mustafa.
234 00:20:14.730 ⇒ 00:20:15.860 Samuel Roberts: You guys can just…
235 00:20:15.860 ⇒ 00:20:16.859 Uttam Kumaran: look at that.
236 00:20:17.010 ⇒ 00:20:23.190 Uttam Kumaran: I also, like… I’m thinking, like, how can we add more linting tests?
237 00:20:23.730 ⇒ 00:20:25.440 Uttam Kumaran: Which I shared…
238 00:20:25.440 ⇒ 00:20:26.450 Samuel Roberts: I saw that, yeah.
239 00:20:26.620 ⇒ 00:20:27.890 Uttam Kumaran: Yeah, so…
240 00:20:28.390 ⇒ 00:20:36.120 Uttam Kumaran: again, I’m just, like, finding things and, like, kind of learning about how other people are dealing with these, like, swarms of stuff that’s coming in, but…
241 00:20:36.570 ⇒ 00:20:40.170 Uttam Kumaran: Maybe, like, this is just, like, linting setup.
242 00:20:41.990 ⇒ 00:20:48.010 Uttam Kumaran: linting setup. Again, this is where I’m almost, like, if, like.
243 00:20:48.300 ⇒ 00:20:57.299 Uttam Kumaran: your, like, AI team and Sam are, like, the people that are, like… you kind of own the house, but then other people kind of live in it, right? So…
244 00:20:57.300 ⇒ 00:20:57.680 Samuel Roberts: Yeah.
245 00:20:57.680 ⇒ 00:21:06.239 Uttam Kumaran: It’s, like, setting the policies, and again, like, I’m okay. Other people are gonna be way worse than me at, like, even understanding, like, any of these words.
246 00:21:06.410 ⇒ 00:21:16.529 Uttam Kumaran: So you gotta think it’s gonna be that bad. So, any point of friction between cursor, build me this feature, and getting it out is gonna be a challenge for people.
247 00:21:16.800 ⇒ 00:21:18.729 Uttam Kumaran: We’re gonna do our best to teach.
248 00:21:19.230 ⇒ 00:21:33.130 Uttam Kumaran: people are having difficulty just even, like, pulling GitHub down. So it’s, like, gonna be worse than you think. So as much of the repo and codecs and cursor that understands the policies to build within itself, within the platform.
249 00:21:33.520 ⇒ 00:21:37.939 Uttam Kumaran: is, like, gonna be way more protection. It’s gonna take us a while to teach people.
250 00:21:39.420 ⇒ 00:21:40.020 Samuel Roberts: Yeah.
251 00:21:40.990 ⇒ 00:21:42.889 Uttam Kumaran: It’s just, like, what we’re seeing, so…
252 00:21:42.890 ⇒ 00:21:45.650 Samuel Roberts: I think, I think, yeah, like, linting, there’s also,
253 00:21:45.760 ⇒ 00:21:50.960 Samuel Roberts: Like, I was having issues with Prettier and stuff, too, which is just, like, code styling, which is, like, kind of linked.
254 00:21:50.960 ⇒ 00:21:51.310 Uttam Kumaran: Yeah.
255 00:21:51.470 ⇒ 00:21:55.419 Samuel Roberts: And then I think there’s probably just, like, more basic tests we want to get in there, too.
256 00:21:56.240 ⇒ 00:21:58.540 Samuel Roberts: Yeah, so, like, some other basic, yeah.
257 00:21:58.560 ⇒ 00:21:59.080 Uttam Kumaran: Yeah.
258 00:21:59.080 ⇒ 00:22:02.419 Samuel Roberts: Yeah, I just, like, smoke test to make sure, like, we can do a health check and stuff, and like…
259 00:22:04.320 ⇒ 00:22:09.160 Samuel Roberts: you know, like, eventually we want to make sure, like, parts of the UI aren’t breaking, but even just the basic, like.
260 00:22:09.570 ⇒ 00:22:11.659 Samuel Roberts: You know, it didn’t break the routing and stuff.
261 00:22:11.950 ⇒ 00:22:16.989 Samuel Roberts: Which I think is probably pretty easy to throw together, like the, you know, the first version of just, like.
262 00:22:17.140 ⇒ 00:22:19.219 Samuel Roberts: Is the server up, you know?
263 00:22:20.120 ⇒ 00:22:20.960 Uttam Kumaran: Yes.
264 00:22:24.010 ⇒ 00:22:24.740 Samuel Roberts: That’s what people might end up.
265 00:22:24.740 ⇒ 00:22:27.660 Uttam Kumaran: If I can also… Yeah, sorry, go ahead.
266 00:22:28.120 ⇒ 00:22:33.669 Mustafa Raja: Yeah, I think Codex can take the screenshots, so that’s something helpful, I believe.
267 00:22:34.340 ⇒ 00:22:39.299 Uttam Kumaran: Yeah, I think this is just gonna be, like, for local development, if we could pull up the browser.
268 00:22:40.090 ⇒ 00:22:45.389 Uttam Kumaran: that… that I think feels like… feels like it works, but really what I’m hoping for… Totally.
269 00:22:45.390 ⇒ 00:22:49.109 Samuel Roberts: When I was doing it yesterday, it took a long time to do some testing of that PR when I.
270 00:22:49.110 ⇒ 00:22:54.279 Uttam Kumaran: Well, what I’m also hoping for is that as part of the PR, Codex can actually…
271 00:22:54.500 ⇒ 00:22:56.860 Uttam Kumaran: Put the screenshot of the feature.
272 00:22:57.250 ⇒ 00:23:02.800 Uttam Kumaran: And, like, I want to get to the point where it generates, like, the feature preview, or the loom.
273 00:23:02.800 ⇒ 00:23:03.879 Samuel Roberts: or something. Yeah, yeah.
274 00:23:04.290 ⇒ 00:23:09.940 Samuel Roberts: That’s what I was… I saw you had, like, for one of the… which… the DEXPR is the one I ran yesterday, trying to do…
275 00:23:10.200 ⇒ 00:23:12.560 Samuel Roberts: Cursor doing some browser testing.
276 00:23:12.880 ⇒ 00:23:15.580 Uttam Kumaran: Yeah, because I needed to test… yeah.
277 00:23:15.830 ⇒ 00:23:25.459 Samuel Roberts: Yeah, I saw there were, like, testing steps, and I was like, oh, follow those steps, and so it did it, but I don’t know if it’s just, like, I don’t have it set up well in the browser, or, like, what was going on.
278 00:23:26.080 ⇒ 00:23:26.730 Uttam Kumaran: Okay.
279 00:23:26.730 ⇒ 00:23:30.610 Samuel Roberts: But it… it was… it took a while to do some pretty basic stuff, I think.
280 00:23:31.690 ⇒ 00:23:32.659 Mustafa Raja: Yeah, and .
281 00:23:32.660 ⇒ 00:23:34.840 Samuel Roberts: That’s in the background, so, you know, it’s not that bad.
282 00:23:34.840 ⇒ 00:23:42.670 Mustafa Raja: When I asked it to, you know, test it, what happened is it took a lot of time, and all it did was, linest.
283 00:23:43.310 ⇒ 00:23:44.139 Mustafa Raja: You know?
284 00:23:45.040 ⇒ 00:23:45.550 Uttam Kumaran: Yeah.
285 00:23:46.630 ⇒ 00:23:48.709 Mustafa Raja: The codecs, not cursor.
286 00:23:48.710 ⇒ 00:23:56.919 Samuel Roberts: Oh, Codex, yeah. Codex, we can probably set up the environment better, because, like, you know, your local cursor already has the environment set up just by default, because, like… Yeah, yeah.
287 00:24:00.460 ⇒ 00:24:07.379 Samuel Roberts: Alright, the one other thing I want to talk about monorepo-wise, now that we need to zoomed transcripts to the vault and stuff, that’s a lot of commits.
288 00:24:08.960 ⇒ 00:24:11.260 Samuel Roberts: I’m just wondering how we’re gonna, like…
289 00:24:11.910 ⇒ 00:24:13.650 Uttam Kumaran: Well, I’ve been batch… it’s been batched.
290 00:24:14.800 ⇒ 00:24:16.790 Samuel Roberts: It has been, but will it be going forward?
291 00:24:18.230 ⇒ 00:24:21.179 Uttam Kumaran: Well, I guess, what’s the problem with a lot of commits?
292 00:24:22.230 ⇒ 00:24:26.189 Samuel Roberts: When someone goes in to look at something in the history.
293 00:24:26.320 ⇒ 00:24:32.239 Samuel Roberts: there’ll just be a lot of noise. And so I’m wondering… there’s ways to, like, squash things?
294 00:24:32.710 ⇒ 00:24:33.320 Uttam Kumaran: Okay.
295 00:24:33.840 ⇒ 00:24:36.780 Samuel Roberts: I’m just not sure how best to do that, because these are going right to Maine, right?
296 00:24:38.220 ⇒ 00:24:41.510 Uttam Kumaran: Yeah, because the intended behavior is, like.
297 00:24:41.860 ⇒ 00:24:44.880 Uttam Kumaran: You got off a meeting, and then it’s there, basically, 30 minutes.
298 00:24:44.880 ⇒ 00:24:49.099 Samuel Roberts: Yeah, no, I get that. That’s why I think that’s the right way to do it. I’m just trying to think, is there a way to, like…
299 00:24:49.880 ⇒ 00:24:57.649 Uttam Kumaran: I guess my question is, though, like, are you going to be accessing the commit history non-programmatically? Like, yes, it is messy, but…
300 00:24:57.850 ⇒ 00:25:01.649 Uttam Kumaran: If you’re just gonna access history via CLI, then you just filter…
301 00:25:02.660 ⇒ 00:25:04.209 Samuel Roberts: Yeah, I mean, I have… I guess it’s…
302 00:25:04.210 ⇒ 00:25:17.659 Uttam Kumaran: I guess I see… yeah, I guess this is also where I’m too dumb to actually get it. Like, I didn’t… I went to… I just got through GitHub, like, 201, I didn’t go past, so… maybe I can ask… I’m totally in for whatever behavior you want to.
303 00:25:17.660 ⇒ 00:25:18.370 Samuel Roberts: Yeah, I’m just…
304 00:25:18.370 ⇒ 00:25:19.060 Uttam Kumaran: forward.
305 00:25:19.630 ⇒ 00:25:21.950 Samuel Roberts: I’m not sure… like.
306 00:25:21.950 ⇒ 00:25:24.809 Gabriel Lam: I mean, can we sort of set a default call to filter?
307 00:25:25.010 ⇒ 00:25:32.360 Gabriel Lam: Or… like, certain PRs that… We’re just like, Let’s just hide this.
308 00:25:32.640 ⇒ 00:25:37.210 Uttam Kumaran: Well, it’s not gonna create a pull request at all, it’s literally gonna write directly to the main branch.
309 00:25:37.550 ⇒ 00:25:39.229 Gabriel Lam: Oh, right, right.
310 00:25:39.470 ⇒ 00:25:40.490 Samuel Roberts: Yeah, cause I don’t…
311 00:25:40.490 ⇒ 00:25:42.219 Uttam Kumaran: I don’t want there to be any gap.
312 00:25:42.220 ⇒ 00:25:46.969 Samuel Roberts: Yeah. No, I think that’s, like, the one thing that should do that, because it’s not… as long as we, like.
313 00:25:47.470 ⇒ 00:25:48.770 Samuel Roberts: Yeah, I think that’s fine.
314 00:25:48.940 ⇒ 00:25:53.599 Uttam Kumaran: Guys, people are, like… we’ve been using this the last 3 days, and it’s been so helpful.
315 00:25:54.120 ⇒ 00:25:54.820 Samuel Roberts: Yeah, yeah, yeah.
316 00:25:54.820 ⇒ 00:26:02.850 Uttam Kumaran: So, the immediate ask I’m gonna get as other people start using it is, like, okay, when are the… when is the next ones coming out? So maybe…
317 00:26:02.850 ⇒ 00:26:03.490 Samuel Roberts: Yeah.
318 00:26:03.820 ⇒ 00:26:05.010 Uttam Kumaran: You guys can just…
319 00:26:05.240 ⇒ 00:26:09.699 Uttam Kumaran: like, I was kind of thinking about, okay, what… what features today can we try to ship?
320 00:26:09.860 ⇒ 00:26:15.120 Uttam Kumaran: for me, it’s like, if we can get the Zoom transcripts, or I… I mean, I can just continue to run it
321 00:26:15.630 ⇒ 00:26:25.180 Uttam Kumaran: like, I can run another batch for the latest on my machine, but, like, if we can get this there, that’d be great. I do think that we need some type of, like, Gmail.
322 00:26:26.450 ⇒ 00:26:32.450 Uttam Kumaran: either MCP or, like, CLI access to… poll emails.
323 00:26:34.870 ⇒ 00:26:35.560 Samuel Roberts: Okay.
324 00:26:36.000 ⇒ 00:26:39.830 Uttam Kumaran: I’m… that’s kind of like an open-ended… Open-ended thing?
325 00:26:39.830 ⇒ 00:26:43.610 Samuel Roberts: the Google access thing that you were doing, Gabe?
326 00:26:43.610 ⇒ 00:26:51.150 Gabriel Lam: I… I’ve only tested, sheets, docs, and Slides.
327 00:26:51.430 ⇒ 00:26:57.060 Gabriel Lam: I should have the APIs for everything else turned on, but I haven’t really had the chance to…
328 00:26:57.060 ⇒ 00:26:57.750 Samuel Roberts: Test that out.
329 00:26:57.750 ⇒ 00:26:58.300 Gabriel Lam: Technically.
330 00:26:58.300 ⇒ 00:27:05.000 Samuel Roberts: Because, like, the nice thing about that is, like, because everyone’s OAuthing, it’s like, it’s just gonna be their email, so it’s not like there’s gonna be any leaking of anything.
331 00:27:06.330 ⇒ 00:27:08.930 Uttam Kumaran: Yeah, this would be, like, super awesome to have.
332 00:27:09.040 ⇒ 00:27:18.670 Uttam Kumaran: like, later today, if you guys can just, like, knock this out, because… yeah. And then… and then really, for this one, we’re gonna need, like, some loom to get to… to show everybody how to do it.
333 00:27:19.840 ⇒ 00:27:20.510 Samuel Roberts: Yeah.
334 00:27:22.750 ⇒ 00:27:28.880 Mustafa Raja: There’s a Google CLI, we can, you know… get, authenticate Gmail with.
335 00:27:29.680 ⇒ 00:27:33.110 Uttam Kumaran: Yeah, I just don’t know if you can access the resource assets.
336 00:27:33.760 ⇒ 00:27:34.240 Samuel Roberts: Yeah.
337 00:27:34.970 ⇒ 00:27:36.339 Uttam Kumaran: I don’t know how it works.
338 00:27:36.970 ⇒ 00:27:42.260 Samuel Roberts: Yeah, the MCP, at least we had to go in and create an app and, like, give it the right access and everything, so it’s…
339 00:27:43.530 ⇒ 00:27:48.790 Samuel Roberts: Okay, we can… we can sort that out a little bit and ask the current… the current,
340 00:27:49.560 ⇒ 00:27:57.609 Samuel Roberts: Google MCP and see if it has Gmail access, and if it does, that’s the solution, I bet, for now, because we can put that in a…
341 00:27:58.020 ⇒ 00:28:03.710 Samuel Roberts: like, top-level cursor folder, probably, for MCPs, right? That way, people will just have it when they download the.
342 00:28:03.710 ⇒ 00:28:06.410 Uttam Kumaran: Yeah, so if you look in… if you look in playbooks.
343 00:28:06.970 ⇒ 00:28:14.530 Uttam Kumaran: Yeah. There’s a whole folder around MCPs, and CLI, MCP, Info.
344 00:28:14.530 ⇒ 00:28:19.159 Samuel Roberts: Right, but I mean, like, if we put a .cursor folder, can you put MCP’s, like, settings in there?
345 00:28:20.150 ⇒ 00:28:26.949 Uttam Kumaran: Well, Cursor can install MCPs in itself, so yeah, you can… like, I can just come in…
346 00:28:26.950 ⇒ 00:28:32.649 Samuel Roberts: Might be a nice way to get people set up on it, where, like, once they download it, and they open it in Cursor, Cursor will see that, and they can OAuth.
347 00:28:33.240 ⇒ 00:28:36.080 Samuel Roberts: We’ll be as, like, as seamless as possible, hopefully.
348 00:28:37.410 ⇒ 00:28:37.910 Uttam Kumaran: Yeah.
349 00:28:37.910 ⇒ 00:28:38.500 Samuel Roberts: Okay.
350 00:28:39.180 ⇒ 00:28:41.860 Gabriel Lam: to the Gmail MCP, the current…
351 00:28:42.320 ⇒ 00:28:48.770 Samuel Roberts: thing we have doesn’t include it, but I can turn on the API and then update it to… Yeah, I turn on API. I was limiting…
352 00:28:49.070 ⇒ 00:28:53.119 Gabriel Lam: Yeah, I was just trying to limit the number of tools we were bringing in, so I did.
353 00:28:53.120 ⇒ 00:28:56.089 Samuel Roberts: No, no, I mean, that’s… you’re… that’s… that’s… you’re not wrong.
354 00:28:56.090 ⇒ 00:29:01.810 Uttam Kumaran: No, we’re not, yeah, we’re just… I’m just, like, basically, like, trying every single thing possible.
355 00:29:02.130 ⇒ 00:29:07.130 Uttam Kumaran: To try to automate what I do, and then being like, okay, this is good, this is a blocker, this is a blocker, this is a blocker.
356 00:29:08.650 ⇒ 00:29:12.240 Samuel Roberts: You want it sending emails, too? Or should we just do, like, read access?
357 00:29:13.620 ⇒ 00:29:16.369 Uttam Kumaran: I would like it to… draft.
358 00:29:19.650 ⇒ 00:29:22.539 Uttam Kumaran: I mean, like, I would… yeah, obviously…
359 00:29:22.540 ⇒ 00:29:22.930 Samuel Roberts: Yeah.
360 00:29:23.100 ⇒ 00:29:26.700 Uttam Kumaran: But I would prefer to not get in the jam where someone loops.
361 00:29:27.010 ⇒ 00:29:28.760 Uttam Kumaran: on stuff? Or, like.
362 00:29:28.760 ⇒ 00:29:29.860 Samuel Roberts: Yeah, yeah.
363 00:29:29.860 ⇒ 00:29:30.470 Uttam Kumaran: F’s up.
364 00:29:30.470 ⇒ 00:29:30.990 Samuel Roberts: Okay, yeah, so it’.
365 00:29:30.990 ⇒ 00:29:31.550 Uttam Kumaran: So…
366 00:29:31.550 ⇒ 00:29:34.409 Samuel Roberts: I don’t know how the API works directly in the setting, but…
367 00:29:34.540 ⇒ 00:29:40.930 Uttam Kumaran: only allow draft… I mean, for this week and next week, it’s not that important, it’s really gonna be, like, probably me.
368 00:29:41.050 ⇒ 00:29:47.939 Uttam Kumaran: But, like, if we can only allow draft creation, not email send, That would be awesome.
369 00:29:48.090 ⇒ 00:29:53.050 Uttam Kumaran: The other ask I have is if we can get a calendar .
370 00:29:53.070 ⇒ 00:29:54.260 Samuel Roberts: API…
371 00:29:54.300 ⇒ 00:29:55.880 Uttam Kumaran: Integration, also?
372 00:29:57.290 ⇒ 00:29:58.090 Uttam Kumaran: Like, G-Cal?
373 00:29:59.810 ⇒ 00:30:02.520 Samuel Roberts: Yeah, I mean, that should be part of that same project, I bet, right?
374 00:30:03.370 ⇒ 00:30:09.780 Uttam Kumaran: Yeah, the ask here is because… well, I’m gonna ask for two things. One is, like, I built a Clockify.
375 00:30:10.150 ⇒ 00:30:11.120 Samuel Roberts: Yeah, I’ve…
376 00:30:11.120 ⇒ 00:30:18.689 Uttam Kumaran: So, it’s gonna be… there’s gonna be two use cases I’m trying to fulfill. One is, like, check out my meetings for the week, and propose
377 00:30:19.300 ⇒ 00:30:21.969 Uttam Kumaran: what… Hours, I should add.
378 00:30:22.360 ⇒ 00:30:25.429 Uttam Kumaran: The second thing I’m gonna do is,
379 00:30:26.230 ⇒ 00:30:30.490 Uttam Kumaran: I’m gonna create a UI that auto basically proposes hours for you.
380 00:30:31.050 ⇒ 00:30:32.910 Samuel Roberts: And then you can just click, click, click.
381 00:30:33.020 ⇒ 00:30:44.340 Uttam Kumaran: So, the two use cases are using cursor to pull CAL events, and then… Then to Clockify…
382 00:30:45.200 ⇒ 00:30:49.409 Uttam Kumaran: creating UI to pull CAL events.
383 00:30:50.050 ⇒ 00:31:00.380 Uttam Kumaran: and give user UI ability to add… 2… Quick add those events.
384 00:31:03.720 ⇒ 00:31:09.920 Uttam Kumaran: My last thing I want to ask is, we’re having people sign in with Google on the platform.
385 00:31:10.090 ⇒ 00:31:15.330 Uttam Kumaran: if I was to go, say, like, I want to auto-filter Or, like.
386 00:31:16.020 ⇒ 00:31:24.590 Uttam Kumaran: I mean, the bigger ask here, I think, is just, like, like… User access control on platform.
387 00:31:24.900 ⇒ 00:31:40.719 Uttam Kumaran: For example, I want to prevent… Most… People from seeing meetings, Bear… Not a part of.
388 00:31:41.400 ⇒ 00:31:42.320 Samuel Roberts: Yeah, yeah.
389 00:31:42.870 ⇒ 00:31:47.930 Uttam Kumaran: So… in my world, I was just gonna send this to Kersher and be like.
390 00:31:48.230 ⇒ 00:31:51.320 Uttam Kumaran: you have access to Google for the…
391 00:31:51.420 ⇒ 00:31:56.549 Uttam Kumaran: sign in, which implies that you have some, like, JWT, or you have whatever.
392 00:31:57.240 ⇒ 00:31:59.749 Uttam Kumaran: Right, you have some, like, user.
393 00:31:59.750 ⇒ 00:32:01.180 Samuel Roberts: They know your identity, yeah.
394 00:32:02.010 ⇒ 00:32:07.179 Uttam Kumaran: Can you… Basically, implement a backend filter
395 00:32:07.570 ⇒ 00:32:15.540 Uttam Kumaran: But this is, yeah, a back-end filter on the database, right, because we don’t want it on the front end, because you can just spoof it. Can you do this on the back end somehow? That’s what I was going to ask.
396 00:32:16.720 ⇒ 00:32:17.100 Samuel Roberts: Yeah.
397 00:32:17.100 ⇒ 00:32:17.579 Uttam Kumaran: But I don’t know.
398 00:32:17.580 ⇒ 00:32:22.409 Samuel Roberts: There is, like, what if you want to share a meeting or something?
399 00:32:22.410 ⇒ 00:32:24.080 Uttam Kumaran: Yeah, we’ll have to enable that, yeah.
400 00:32:24.320 ⇒ 00:32:26.369 Samuel Roberts: Yeah, cause, I mean, yeah, I mean, that’s sort of…
401 00:32:27.080 ⇒ 00:32:30.780 Uttam Kumaran: Well, we’ll have to enable, like, basically, like, a request access flow.
402 00:32:30.780 ⇒ 00:32:31.460 Samuel Roberts: Yeah.
403 00:32:32.990 ⇒ 00:32:36.110 Uttam Kumaran: Yeah, but again, like, I will… I will, kind of.
404 00:32:36.670 ⇒ 00:32:40.200 Uttam Kumaran: I’m outlining just, like, what features I think I would try to push.
405 00:32:40.800 ⇒ 00:32:41.370 Samuel Roberts: Yeah.
406 00:32:41.370 ⇒ 00:32:44.409 Uttam Kumaran: And then we… most people, I think we would… like, I just want to start…
407 00:32:45.860 ⇒ 00:32:47.730 Uttam Kumaran: Being a bit more aware.
408 00:32:47.990 ⇒ 00:32:50.449 Uttam Kumaran: So this is, like, kind of something I would try to push.
409 00:32:51.280 ⇒ 00:32:54.810 Samuel Roberts: Yeah, I think that’s good. I think the Teams table is in there, too.
410 00:32:56.430 ⇒ 00:33:01.410 Samuel Roberts: I think I still need to double-check that it actually runs on a cadence, because I manually ran it the other day.
411 00:33:01.430 ⇒ 00:33:02.569 Mustafa Raja: Oh, yes, I really…
412 00:33:02.750 ⇒ 00:33:04.500 Samuel Roberts: I didn’t realize. Yeah.
413 00:33:04.500 ⇒ 00:33:12.209 Mustafa Raja: I think the only caveat here would be, that the Zoom table really has names of the people rather than emails of the people, you know?
414 00:33:12.370 ⇒ 00:33:12.980 Mustafa Raja: It’s.
415 00:33:12.980 ⇒ 00:33:15.170 Gabriel Lam: As emails now, we made that on Facebook.
416 00:33:15.170 ⇒ 00:33:16.809 Samuel Roberts: Act, yeah. Thanks, Casey.
417 00:33:16.810 ⇒ 00:33:17.389 Uttam Kumaran: That’s nice.
418 00:33:17.390 ⇒ 00:33:18.260 Mustafa Raja: Okay, okay, yeah.
419 00:33:18.260 ⇒ 00:33:20.730 Samuel Roberts: Yeah, we ran into that problem the other day, and we just…
420 00:33:21.040 ⇒ 00:33:25.999 Samuel Roberts: Casey Bank, so now there are emails stored, because that was exactly what we were noticing.
421 00:33:26.130 ⇒ 00:33:28.670 Samuel Roberts: Trying to do the linear stuff.
422 00:33:28.990 ⇒ 00:33:33.949 Samuel Roberts: So now there are emails, so we can cross-reference emails with the user.
423 00:33:36.470 ⇒ 00:33:46.520 Samuel Roberts: that is signed in. That shouldn’t be crazy in its most basic form. Yeah. The access… requesting access and stuff might be a little more, but if you want to just prevent right now, we could probably do that pretty quickly.
424 00:33:46.520 ⇒ 00:33:48.359 Mustafa Raja: Yeah, that’s super easy then, right?
425 00:33:49.490 ⇒ 00:33:50.060 Samuel Roberts: Yeah.
426 00:33:50.670 ⇒ 00:33:51.649 Mustafa Raja: Okay, I need to hop.
427 00:33:52.310 ⇒ 00:33:52.900 Uttam Kumaran: Okay.
428 00:33:52.900 ⇒ 00:33:54.990 Samuel Roberts: Okay, yeah, sorry, yeah, I didn’t realize we were going so long.
429 00:33:55.360 ⇒ 00:33:56.420 Samuel Roberts: But yeah, I think…
430 00:33:59.730 ⇒ 00:34:05.160 Samuel Roberts: I think that’s doable, and probably worth having in place right now, like, ASAP.
431 00:34:06.510 ⇒ 00:34:13.199 Uttam Kumaran: Okay. Yeah, I’ll probably… I have, like, a big block for this ABC pitch later, but I will probably be on…
432 00:34:13.780 ⇒ 00:34:16.139 Uttam Kumaran: In the evening, again, working on some of these.
433 00:34:16.960 ⇒ 00:34:23.030 Samuel Roberts: Okay, ABC Pitch, is that related to the Abade stuff I was doing, or is this different?
434 00:34:23.590 ⇒ 00:34:26.689 Uttam Kumaran: No, for the discovery… for the discovery.
435 00:34:26.690 ⇒ 00:34:32.510 Samuel Roberts: Okay, okay, I was just making sure, because I’m still in the middle of that, I didn’t know what… okay, yeah. I made some good progress yesterday with it, though, it’s hopefully gonna do.
436 00:34:32.510 ⇒ 00:34:33.209 Uttam Kumaran: Oh, sick, okay.
437 00:34:33.219 ⇒ 00:34:33.939 Samuel Roberts: I don’t know, I got the key.
438 00:34:33.940 ⇒ 00:34:35.820 Uttam Kumaran: Well, yeah, I mean, if you actually have…
439 00:34:35.969 ⇒ 00:34:38.170 Uttam Kumaran: Do you have any progress? Like, I can…
440 00:34:39.340 ⇒ 00:34:43.520 Samuel Roberts: I… I got it to a point when I could…
441 00:34:44.960 ⇒ 00:34:53.039 Samuel Roberts: yeah, I just had to give it the keys last night, and then run it, and so I did that this morning, and it didn’t work, so I’m still going with Codex a little bit.
442 00:34:53.920 ⇒ 00:35:00.819 Samuel Roberts: Okay. But it implemented the plan, we’re just trying to see if it’s, you know, what the little edge cases are, where it’s not pulling stuff, but I’ll hopefully have more updates on that later anyway.
443 00:35:01.710 ⇒ 00:35:02.240 Uttam Kumaran: Okay.
444 00:35:02.900 ⇒ 00:35:03.750 Samuel Roberts: One way or another.
445 00:35:04.540 ⇒ 00:35:10.359 Samuel Roberts: And then, Gabe, for the Google MCP stuff, we had issues running it in Railway, right?
446 00:35:11.180 ⇒ 00:35:14.270 Gabriel Lam: Yeah, I… would appreciate.
447 00:35:14.700 ⇒ 00:35:17.900 Gabriel Lam: Like, work session or, like, a review on that.
448 00:35:17.900 ⇒ 00:35:19.369 Samuel Roberts: Okay, I was gonna say.
449 00:35:19.370 ⇒ 00:35:23.550 Gabriel Lam: It might just mean me being stupid, and probably is, but .
450 00:35:23.550 ⇒ 00:35:29.179 Samuel Roberts: It’s… there’s always something with that, you know what I mean? Especially deployment stuff, like, especially deployment with Google secrets and, like.
451 00:35:29.300 ⇒ 00:35:41.740 Samuel Roberts: the right domain, and yeah, there’s all kinds of rabbit holes that can trip you up there. Yeah, let’s… if you want to get something on the calendar for, like, an hour, we could probably get that deployed, because I think having that up there will be…
452 00:35:42.120 ⇒ 00:35:44.020 Samuel Roberts: Much better than people running it locally.
453 00:35:44.940 ⇒ 00:35:46.830 Gabriel Lam: In terms of just simplicity.
454 00:35:50.000 ⇒ 00:35:50.920 Samuel Roberts: Cool.
455 00:35:51.150 ⇒ 00:35:52.790 Samuel Roberts: Anything else?
456 00:35:58.070 ⇒ 00:35:58.960 Samuel Roberts: Yeah.
457 00:35:59.070 ⇒ 00:35:59.770 Samuel Roberts: Okay.
458 00:36:02.500 ⇒ 00:36:03.170 Samuel Roberts: Okay.
459 00:36:03.170 ⇒ 00:36:03.820 Uttam Kumaran: I gotta jump.
460 00:36:03.820 ⇒ 00:36:06.369 Samuel Roberts: Okay, yeah, sounds good. Alright.
461 00:36:06.370 ⇒ 00:36:07.690 Uttam Kumaran: Okay, thank you.
462 00:36:07.690 ⇒ 00:36:09.289 Samuel Roberts: See y’all later, then. Thanks, guys.
463 00:36:09.670 ⇒ 00:36:10.410 Samuel Roberts: Yep.
464 00:36:10.410 ⇒ 00:36:11.180 Casie Aviles: Thank you.