Meeting Title: Uttam-Kumaran’s-Personal-Meeting-Room Date: 2024-03-11 Meeting participants: Agustin Bergoglio, Ryan Luke Daque, Patrick Trainer, Uttam Kumaran


WEBVTT

1 00:01:04.379 00:01:05.379 Ryan Luke Daque: Hello.

2 00:01:09.660 00:01:10.550 Uttam Kumaran: hey.

3 00:01:10.980 00:01:12.730 Agustin Bergoglio: guys! Hello!

4 00:01:15.720 00:01:19.390 Ryan Luke Daque: I guess let’s just so. Wait a bit for Patrick.

5 00:02:22.790 00:02:24.299 Ryan Luke Daque: I just paid it in.

6 00:02:36.310 00:02:41.749 Agustin Bergoglio: Sorry I have a question, and we have the a client that’s called pull parts to go right?

7 00:02:42.040 00:02:42.980 Ryan Luke Daque: Right?

8 00:02:43.310 00:02:46.010 Agustin Bergoglio: So you see, I’m the only client

9 00:02:46.640 00:02:48.560 Agustin Bergoglio: that uses Dvt.

10 00:02:49.780 00:02:59.920 Ryan Luke Daque: No, we also have like asset link. And II believe the the others as well. Right? Yeah, we have 2 clients that are currently we are managing their DVD projects.

11 00:03:00.140 00:03:03.759 Uttam Kumaran: And then we have another client Ampla.

12 00:03:03.820 00:03:05.889 Uttam Kumaran: that they have Dvt internally

13 00:03:05.970 00:03:13.369 Uttam Kumaran: but basically, you know any client where we’re pretty much doing data modeling for we are suggesting Dvp.

14 00:03:13.540 00:03:23.879 Uttam Kumaran: and the kind of the the onus of this project, was it? Basically, it takes us, like probably a week, to get everything set up and aligned. And

15 00:03:23.890 00:03:28.159 Uttam Kumaran: I think we can pretty much do that in a day. And

16 00:03:28.230 00:03:40.359 Uttam Kumaran: if clients need Dvt best practices, I would like some documentation to provide for them. So basically for us to get a line on, all of our projects will be structured the same way that way. Anyone can go and work on anything, and

17 00:03:40.620 00:03:47.449 Uttam Kumaran: the focus isn’t on like aware of something. And then, second, you know, to help all the other clients make sure that their Dbt project serve

18 00:03:47.580 00:03:49.210 Uttam Kumaran: or formatted the right way. So

19 00:03:50.300 00:03:51.100 Ryan Luke Daque: nice?

20 00:03:53.420 00:03:56.189 Uttam Kumaran: Yeah. So yeah, you could probably kick it off. Brian.

21 00:03:56.290 00:04:01.219 Ryan Luke Daque: Yeah, sure. So yeah, I created this document so that we can

22 00:04:01.680 00:04:06.229 Ryan Luke Daque: decide or like agree on a standard way to

23 00:04:06.300 00:04:10.779 Ryan Luke Daque: structure our dB projects, especially like the models and stuff like that. So currently.

24 00:04:11.010 00:04:18.600 Ryan Luke Daque: we have our DVD project set up like for sample for full parts to go. It’s mostly

25 00:04:19.300 00:04:30.420 Ryan Luke Daque: on a per source level the our folders under the models are like per source. and we don’t even have like staging models to do like the pre.

26 00:04:30.630 00:04:36.729 Ryan Luke Daque: like naming ids, or like changing schemas and stuff like that like

27 00:04:36.820 00:04:39.880 Ryan Luke Daque: breed transformation. And maybe it’s like

28 00:04:40.270 00:04:49.150 Ryan Luke Daque: it’s like one of Dbt suggested best practices to have staging models so that we can like have a consistent schemas for ids, for example, or like.

29 00:04:49.300 00:04:53.280 Ryan Luke Daque: we can rename stuff before

30 00:04:53.680 00:05:00.370 Ryan Luke Daque: the final models, right? So we don’t have to keep on renaming them every time we use those sources and stuff like that.

31 00:05:01.140 00:05:08.580 Ryan Luke Daque: So yeah, basically the folder structure that I’m I’m trying to have everyone

32 00:05:08.740 00:05:18.859 Ryan Luke Daque: agree or like, see if there’s anything else we can change is basically like this. This is also Dbt’s suggestive best practice in terms of like

33 00:05:19.030 00:05:25.690 Ryan Luke Daque: older structure, we have staging layer model per source.

34 00:05:25.860 00:05:31.549 Ryan Luke Daque: And then we have intermediate models. If ever we need to like combine sources from different.

35 00:05:32.250 00:05:33.960 Ryan Luke Daque: I mean different.

36 00:05:34.220 00:05:39.080 Ryan Luke Daque: we need to combine staging models from different sources to create some

37 00:05:39.420 00:05:41.669 Ryan Luke Daque: calculations and stuff like that.

38 00:05:41.900 00:05:47.699 Ryan Luke Daque: And then we have the final march models which would be the models that will be used for

39 00:05:48.180 00:05:52.759 Ryan Luke Daque: the business intelligence and analytics visual boards and stuff like that.

40 00:05:53.010 00:06:04.910 Ryan Luke Daque: And this would be like divided into whatever category that specific model is like, whether it’s for marketing, it’s for sales, or whatever customers and stuff like that.

41 00:06:05.980 00:06:07.620 Ryan Luke Daque: So yeah, for

42 00:06:07.920 00:06:15.259 Ryan Luke Daque: parts to go. For example, this would potentially be like how the staging models would look like

43 00:06:15.680 00:06:23.910 Ryan Luke Daque: it’s still going to be per source. But then we’ll do all the initial transformations here, like renaming the fields changing data types.

44 00:06:23.940 00:06:27.599 Ryan Luke Daque: even standardizing names for ids. Let’s say,

45 00:06:29.270 00:06:36.379 Ryan Luke Daque: like, we we can like use snake case for everything. Some of them are probably

46 00:06:36.540 00:06:51.229 Ryan Luke Daque: what do you call that like? No underscores, and just and stuff like that? So we can. We can standardize all the names there. I haven’t added it. But yeah, like.

47 00:06:51.350 00:06:52.790 Ryan Luke Daque: we can use.

48 00:06:53.580 00:06:58.290 Uttam Kumaran: Okay, great, great, great. There’s something about standardizing naming conventions.

49 00:06:59.210 00:07:06.130 Ryan Luke Daque: Yeah. Yeah. I’ll add, okay, okay, great, yeah, yeah, this is actually super great. Because

50 00:07:06.680 00:07:07.380 Uttam Kumaran: with

51 00:07:07.670 00:07:22.000 Uttam Kumaran: don’t like making these decisions. And I would rather just have us have rules that I don’t really have to like cause I’m not. This is where, like you all want my creativity. So we all. If we’re doing like Snake case and everything from other stores, then

52 00:07:22.040 00:07:29.940 Uttam Kumaran: I’m happy to do that. And then we can actually make sure that things are like that. The only other thing I also wanted to note down is materialization.

53 00:07:30.230 00:07:38.430 Uttam Kumaran: So I wanna know, like in which categories there’s like optional versus course materializations.

54 00:07:38.460 00:07:43.240 Uttam Kumaran: And then that way also, we don’t have to do materializations based on

55 00:07:43.280 00:07:47.100 Uttam Kumaran: individual files. We could do it almost at like the sub folder level.

56 00:07:47.470 00:07:57.750 Uttam Kumaran: I think, for now would be like all things in March. get materialized, and then everything else is on a case by case basis, basically

57 00:07:58.810 00:08:03.580 Ryan Luke Daque: for the most part. Yeah, yeah, big views.

58 00:08:04.730 00:08:20.150 Ryan Luke Daque: And yeah, March would be like tables or like partition tables depending like on what we want to do like. We can also like do incremental models for some if needed, like for for very large datasets and stuff like that. We can. We can make them incremental models and stuff.

59 00:08:20.980 00:08:22.160 Ryan Luke Daque: Yeah.

60 00:08:22.580 00:08:26.040 Ryan Luke Daque: yeah, here, I’ll just take more than that.

61 00:08:26.290 00:08:31.099 Patrick Trainer: Another another good thing to point out is like what you can

62 00:08:31.320 00:08:37.540 Patrick Trainer: like import from. So we have, like the sources, the intermediate, and then the final stage.

63 00:08:37.559 00:08:44.900 Patrick Trainer: like a final stage models should only selector import from

64 00:08:45.770 00:08:56.749 Patrick Trainer: yeah, from from the from like a stage 2. And then, like we, we should never have, like a source, go into like a stage one. Go into a stage 3

65 00:08:57.800 00:09:08.990 Patrick Trainer: it like a source model. Go into yeah, and or or the other way around so just like maintaining. But linearity there is right

66 00:09:09.630 00:09:17.090 Uttam Kumaran: that makes sense. Is there any? Is there any way we can like enforce these or like? And are there any observability on like

67 00:09:17.410 00:09:21.400 Uttam Kumaran: these sorts of relationships right now, is there any tools that allow that?

68 00:09:21.420 00:09:31.150 Uttam Kumaran: Because, like this is where we’re gonna start, that we’re gonna set up divergence over time. So I wanna make sure we have some cadence about looking back at these rules and making sure things are

69 00:09:31.210 00:09:38.090 Patrick Trainer: right nothing that I know of. Like we’ve

70 00:09:38.220 00:09:41.059 Patrick Trainer: I’ve had past teams where we’ve enforced

71 00:09:41.180 00:09:44.580 Patrick Trainer: like naming conventions through

72 00:09:45.300 00:09:52.120 Patrick Trainer: like git hooks and that sort of stuff, but that really mostly deals with with files

73 00:09:52.410 00:10:00.810 Patrick Trainer: or like the files themselves. We could write some sort of parsing script that, like Grept through

74 00:10:01.670 00:10:09.569 Patrick Trainer: the Directory looking for like the curly, curly source. I wonder if people mesh, we can do something with that, too.

75 00:10:10.630 00:10:11.899 Patrick Trainer: Yeah, maybe so.

76 00:10:12.040 00:10:20.660 Patrick Trainer: maybe so. there’s a couple of ways to that. We could probably

77 00:10:20.880 00:10:25.880 Patrick Trainer: figure it out. I don’t think it’s too difficult, but it’d be it’d be loose, and

78 00:10:26.160 00:10:28.570 Patrick Trainer: yeah, more or less just for us

79 00:10:28.880 00:10:31.459 Patrick Trainer: could be a cool spin off

80 00:10:31.570 00:10:33.050 Patrick Trainer: project. The

81 00:10:33.350 00:10:44.849 Uttam Kumaran: yeah. No, basically cause. The things with all these rules is that like again, I’ve been on the on the end of creating these for a few years now. And basically it’s one am I creating? But I also wanna

82 00:10:44.920 00:10:56.109 Uttam Kumaran: every step of the way on these sorts of things. I wanna try to automate and have automatic enforcement, or at least on Prs, that’s the only way this is gonna work.

83 00:10:56.260 00:10:57.940 Uttam Kumaran: because we’re all gonna

84 00:10:58.170 00:11:08.579 Uttam Kumaran: we’re all gonna have moments where we like, just miss miss them. So that’s what I would suggest. Maybe we can note at the end of this document.

85 00:11:08.760 00:11:15.850 Uttam Kumaran: Ryan, we’re just discussing that. I think the big. We could maybe have a section at the end of this talk. We’re just like to do, or further explorations.

86 00:11:15.900 00:11:17.309 Uttam Kumaran: I wanted to have

87 00:11:17.480 00:11:25.120 Uttam Kumaran: some sort of enforcement of these rules. and some sort of understanding of like our coverage. On adhering to them.

88 00:11:25.170 00:11:26.200 Uttam Kumaran: I don’t know what

89 00:11:26.690 00:11:30.020 Uttam Kumaran: what way we kind of handle that now, but

90 00:11:30.120 00:11:33.080 Uttam Kumaran: it would be really, really great to have something, because

91 00:11:33.090 00:11:38.060 Ryan Luke Daque: otherwise this document is, gonna sit there. But I want some sort of automatic system to be able to

92 00:11:38.210 00:11:39.429 Uttam Kumaran: force this

93 00:11:39.650 00:11:45.870 Uttam Kumaran: and it’s a car to get into the habit of like when we set up these documentations, or, for example, when we do

94 00:11:45.940 00:11:51.100 Uttam Kumaran: a similar effort on tests like, I want some sort of coverage, basically.

95 00:11:52.660 00:11:54.340 Ryan Luke Daque: yeah, that makes sense.

96 00:11:56.620 00:12:06.970 Agustin Bergoglio: Sorry. I have a question how the test coverage in Dvt. I’m not sure I have tested it on other backhand perfect.

97 00:12:07.320 00:12:08.280 Agustin Bergoglio: but not

98 00:12:08.560 00:12:15.560 Uttam Kumaran: we don’t have any metric for test coverage now. But basically, I said that it would be great to have

99 00:12:15.880 00:12:19.069 Uttam Kumaran: couple of default tests for every model that we push.

100 00:12:19.100 00:12:22.700 Ryan Luke Daque: and then we can have some level of optional tests

101 00:12:23.150 00:12:27.370 Uttam Kumaran: on on any effort. I definitely want every model to have it attached.

102 00:12:27.880 00:12:35.450 Uttam Kumaran: but I think for backend and stuff there are really like. there are pretty good methods for doing

103 00:12:35.860 00:12:37.650 Uttam Kumaran: test coverage. Right? Like.

104 00:12:39.270 00:12:49.060 Ryan Luke Daque: yeah, I think we can actually use the lies this elementary Dbt package. To do that I’ll have to do some further research, though I think I’ve

105 00:12:49.720 00:12:54.900 Ryan Luke Daque: tried this before, but I can’t quite remember, like what what we can do

106 00:12:55.040 00:13:02.650 Ryan Luke Daque: with this. II think, yeah, I think there’s a possibility of like being able to monitor, test and stuff like that

107 00:13:03.100 00:13:06.920 Ryan Luke Daque: and do like snapshots of both rooms and stuff.

108 00:13:07.020 00:13:17.380 Ryan Luke Daque: Okay? Yeah. And and again, I think of all these is like, you know, we’re buying. We’re buying back our time in the future when

109 00:13:17.680 00:13:20.740 Uttam Kumaran: when these comes up as blogs, basically so

110 00:13:20.780 00:13:37.519 Uttam Kumaran: kind of like, partly insurance is partly like saving us headache later. So and again, I think, as we. We’ve done this once, twice, and of course we’ve all done this a couple of times outside of Brainforge. Now we can really get good. Hey? Cool! If we were to do next. Dbt, project.

111 00:13:37.550 00:13:39.230 Uttam Kumaran: let’s see if we can have it.

112 00:13:39.340 00:13:49.039 Uttam Kumaran: I have like 5 times less bugs. Take 5 times quicker, you know. That’s supportive sort of stuff that I always want to kind of be thinking about. Those sort of step function improvements

113 00:13:49.690 00:13:50.909 Ryan Luke Daque: makes sense.

114 00:13:52.860 00:14:00.509 Ryan Luke Daque: Yeah, but I’ll I’ll do some research on this. I think this even has like a normally detection for tests and stuff like that. You can see

115 00:14:03.260 00:14:08.130 Ryan Luke Daque: if there’s any like weird stuff going on for the tests. anyway.

116 00:14:09.200 00:14:20.030 Ryan Luke Daque: But yeah, that’s what I’ve that’s what I have so far. I’ll I’ll add in also the yeah, the materialization like for the exploration. How we can like automate these

117 00:14:20.320 00:14:23.550 Ryan Luke Daque: to make sure we’re we’re actually doing them.

118 00:14:24.580 00:14:28.080 Patrick Trainer: yeah, this looks good.

119 00:14:29.410 00:14:37.190 Ryan Luke Daque: What about your thoughts on like March models referencing other March models? With that?

120 00:14:37.270 00:14:38.469 Ryan Luke Daque: That would be like.

121 00:14:38.860 00:14:41.170 Patrick Trainer: yeah, I like, I think that’s fun

122 00:14:42.060 00:14:45.600 Patrick Trainer: like, because it’s in the same layer. I wouldn’t have

123 00:14:46.210 00:14:50.210 Patrick Trainer: like a an intermediate layer reference, a mart layer.

124 00:14:50.370 00:14:52.349 Patrick Trainer: But like the other way around, it’s fun.

125 00:14:52.760 00:14:53.440 Uttam Kumaran: Yeah.

126 00:14:54.170 00:14:55.800 Ryan Luke Daque: Yeah. Makes sense. Then.

127 00:14:56.680 00:14:57.480 Ryan Luke Daque: okay.

128 00:14:57.610 00:15:01.089 Ryan Luke Daque: yeah. If, like, everybody agrees on this, then we can.

129 00:15:01.310 00:15:02.120 Agustin Bergoglio: Yes.

130 00:15:02.250 00:15:06.560 Ryan Luke Daque: we can start using this moving forward. But I don’t know if we.

131 00:15:06.630 00:15:13.990 Ryan Luke Daque: because if we need to do that for pool parts, it’s gonna be a quite a huge yeah, it’s it’s it’s going to be a pretty big lift.

132 00:15:14.040 00:15:15.070 Ryan Luke Daque: Yeah.

133 00:15:15.190 00:15:18.589 Patrick Trainer: I think one of the like.

134 00:15:18.760 00:15:25.070 Patrick Trainer: I mean, it’s tedious. But one of the best ways to tackle this now is to

135 00:15:25.370 00:15:37.179 Patrick Trainer: kind of like section off everything, and then distill those sections and tickets and then create subtickets based off of that like, it becomes a lot. But like if we

136 00:15:37.340 00:15:41.399 Patrick Trainer: chunk it down into very like small bite size pieces

137 00:15:41.450 00:15:45.800 Patrick Trainer: then it then like, that’s really the only way that

138 00:15:45.860 00:15:57.029 Patrick Trainer: I’ve been able to get through like big project migrations like this. Yeah, I think like we, I don’t. I don’t think because of how closely linked the tables are to light dash, and I’ve asked them

139 00:15:57.100 00:16:09.840 Uttam Kumaran: about how to make those changes. It’s kind of nightmare. We probably won’t be able to enforce the model naming conventions, but I definitely think it’s not too big of a left to enforce the

140 00:16:10.100 00:16:12.380 Uttam Kumaran: mark’s intermediary staging.

141 00:16:12.400 00:16:14.340 Patrick Trainer: I think there’s a lot of file moving.

142 00:16:14.470 00:16:15.660 Uttam Kumaran: And then.

143 00:16:15.930 00:16:21.529 Uttam Kumaran: like, basically, almost everything is getting materialized now. So we could do that.

144 00:16:21.550 00:16:23.319 Uttam Kumaran: The other thing.

145 00:16:23.700 00:16:32.020 Uttam Kumaran: Ryan, you just you just didn’t have in your document now is that snowflake structure? So it’d be great to.

146 00:16:32.140 00:16:41.789 Uttam Kumaran: I guess we have everything. Those are just the steamer names, Marz interior. But I guess just one thing you can say is, we’re gonna name everything what the database names are going to be

147 00:16:41.920 00:16:45.729 Uttam Kumaran: that’s just Dbt or dbt, prod, or something like that.

148 00:16:45.760 00:16:51.970 Ryan Luke Daque: Yeah, I’m just gonna bring that up as well. But I haven’t like added it into the document, because

149 00:16:52.130 00:16:53.349 Ryan Luke Daque: I’m not sure

150 00:16:53.400 00:16:56.019 Ryan Luke Daque: this point like what would be the best way to

151 00:16:56.460 00:16:58.500 Ryan Luke Daque: do that. But

152 00:16:58.790 00:17:13.670 Patrick Trainer: there. There’s also ways that we can like, based off of the yeah, based off the profile or like the folder structure. Yeah, you can dynamically name schemas or databases

153 00:17:14.300 00:17:20.859 Patrick Trainer: that is like all that entails, is kind of like a a couple of

154 00:17:21.109 00:17:29.200 Patrick Trainer: like macros that override some of the internals, but like it it. They they work really. Well, that’s how I’ve always done it.

155 00:17:29.310 00:17:39.080 Ryan Luke Daque: I think we can even do it in a pro in the project. Gamble.

156 00:17:40.110 00:17:43.420 Agustin Bergoglio: regarding the snowflake

157 00:17:43.640 00:17:51.280 Ryan Luke Daque: structure, though, should I was like thinking, maybe we have one database just for all the raw sources like

158 00:17:51.500 00:17:57.450 Ryan Luke Daque: maybe brain, whatever like cause. Currently we have 5 grand database. This is raw sources

159 00:17:57.740 00:18:06.180 Ryan Luke Daque: with all the tables here, and then we have another one for our specific internal data pipeline that I was thinking created.

160 00:18:06.600 00:18:18.830 Ryan Luke Daque: And then. yeah, like, we also have Nextsla, this is another source. So maybe we can have just one database for sources. And then like, yeah. And like, maybe dbt.

161 00:18:19.780 00:18:21.490 Ryan Luke Daque: a fraud

162 00:18:21.670 00:18:25.110 Ryan Luke Daque: database for production and like one for

163 00:18:25.820 00:18:26.620 Patrick Trainer: right?

164 00:18:27.360 00:18:28.220 Uttam Kumaran: Yeah.

165 00:18:31.160 00:18:37.320 Uttam Kumaran: okay, so let’s just like, keep. Maybe keep this in mind for now. And so the one thing ticket maybe you can make, or one issue you can make

166 00:18:37.600 00:18:42.270 Uttam Kumaran: Ryan and cool parts is just having a ticket for

167 00:18:42.320 00:18:46.510 Uttam Kumaran: actually making that change to the repo structure.

168 00:18:46.630 00:18:51.750 Uttam Kumaran: And you can just put in the backlog, and we can make sure we triage and have that subcast out.

169 00:18:51.880 00:18:53.880 Uttam Kumaran: And we can split up the work on that one.

170 00:18:54.360 00:18:56.170 Agustin Bergoglio: But basically just like applying.

171 00:18:56.430 00:18:59.890 Uttam Kumaran: applying this new convention to that repo

172 00:19:00.130 00:19:01.090 Ryan Luke Daque: and then

173 00:19:01.190 00:19:03.730 Uttam Kumaran: doing a similar one for asset. Mac

174 00:19:05.870 00:19:08.710 Ryan Luke Daque: Gotcha sounds good.

175 00:19:11.140 00:19:11.950 Ryan Luke Daque: Yeah.

176 00:19:12.100 00:19:17.449 Ryan Luke Daque: I think that’s that’s about it. Anything else? Anyone wants to discuss or anything.

177 00:19:20.020 00:19:21.190 Patrick Trainer: That’s it for me.

178 00:19:21.690 00:19:22.610 Ryan Luke Daque: Yes, I will.

179 00:19:23.660 00:19:25.140 Agustin Bergoglio: Yeah.

180 00:19:25.420 00:19:32.499 Ryan Luke Daque: alright. Can we add suggestions to this documentation? Add comments, or whatever.

181 00:19:33.270 00:19:35.240 Ryan Luke Daque: Okay, great.

182 00:19:35.400 00:19:37.870 Patrick Trainer: one good thing, too, is to

183 00:19:38.090 00:19:41.509 Patrick Trainer: like. We put this document

184 00:19:41.650 00:19:53.940 Patrick Trainer: in? The repo asked like a like a mark down file. And then, as we change that, we’ll have to get history for it, too. I’ve always found that helpful.

185 00:19:54.280 00:19:58.070 Uttam Kumaran: Oh, yeah, yeah, maybe, Ryan, you can literally just like, put this in

186 00:19:58.840 00:20:01.479 Uttam Kumaran: strategy routine how to do converts a mark down.

187 00:20:02.510 00:20:06.750 Uttam Kumaran: And then let’s just like, maybe once Augustine’s changes are in, we can just

188 00:20:06.820 00:20:08.879 Uttam Kumaran: we could just confirm this. And

189 00:20:08.950 00:20:15.929 Patrick Trainer: sorry. Are you talking about this Dvt documentation.

190 00:20:16.690 00:20:22.700 Patrick Trainer: yeah, or or like any and all documentation or or like how to use or

191 00:20:23.030 00:20:28.040 Patrick Trainer: work like work workflows, cookbooks, whatever you wanna call them

192 00:20:28.610 00:20:32.050 Patrick Trainer: just for items like a markdown or plain tech stock.

193 00:20:32.060 00:20:37.589 Patrick Trainer: and like, commit that to the repo to the I’ve internal repo is good.

194 00:20:37.670 00:20:39.100 Patrick Trainer: and then

195 00:20:39.350 00:20:40.919 we can create

196 00:20:41.030 00:20:43.700 Patrick Trainer: or or like as we change it and update it.

197 00:20:43.890 00:20:47.750 Patrick Trainer: we’ll have that get history. But then we can also link issues

198 00:20:47.790 00:20:53.970 Patrick Trainer: to it as well and track kind of like our conversations around like thought processes.

199 00:20:54.110 00:21:02.390 Patrick Trainer: And like, when we’d still all those and issues, you then get the and combine that with, get history, you get like the

200 00:21:02.640 00:21:06.039 Patrick Trainer: the context of what was being said as well as like

201 00:21:06.120 00:21:13.819 Patrick Trainer: what was being said as well. And then what changed when? So you get like a really good view?

202 00:21:14.090 00:21:15.569 Agustin Bergoglio: I have a question.

203 00:21:15.630 00:21:20.709 Agustin Bergoglio: yeah, yeah, go for it. Sorry II have worked with the bucket

204 00:21:20.720 00:21:25.660 Agustin Bergoglio: but we were able to link like a ticket

205 00:21:25.850 00:21:28.920 Patrick Trainer: to the to the branch. Right?

206 00:21:29.020 00:21:37.690 Patrick Trainer: Yup, can we do that? We’d have? Yes, yes, yeah. You can do that. And I, there’s actually like, if you’re on an issue.

207 00:21:37.940 00:21:39.960 Patrick Trainer: Like any

208 00:21:40.410 00:21:42.740 Patrick Trainer: issue. There’s

209 00:21:42.920 00:21:49.500 Patrick Trainer: like a a button on it. I think it’s on the right side. It’s you can create a branch in a role.

210 00:21:50.080 00:21:55.599 Patrick Trainer: Take like the issue, name and number and and link that to your branch.

211 00:21:55.630 00:22:02.820 Patrick Trainer: You can also reference issues in commits so if you put just like the hashtag

212 00:22:03.570 00:22:08.579 Patrick Trainer: issue number that’ll link your commit. Yeah, see? Over there in development. There’s the

213 00:22:08.840 00:22:10.220 Patrick Trainer: create a branch. Yeah.

214 00:22:10.440 00:22:13.980 Ryan Luke Daque: That’ll do that.

215 00:22:14.000 00:22:17.639 Patrick Trainer: Then you can. Also, you can also reference

216 00:22:18.450 00:22:27.609 Patrick Trainer: like issues to to pull requests. There’s a lot of references that you can throw around everywhere. I’ll throw. I’ll throw some documentation in here.

217 00:22:27.700 00:22:28.790 Patrick Trainer: but

218 00:22:29.060 00:22:45.629 Patrick Trainer: it makes it super powerful, especially when it goes into it, and it creates with the project tractor. It’s like, if you close an issue or have a pull request, pull, request that closes as an issue. It’ll move your project tile across the board

219 00:22:45.840 00:22:48.909 Patrick Trainer: and so like, you don’t have to

220 00:22:48.930 00:22:56.340 Patrick Trainer: eventually like you can have a bunch of things that cascade and move, move everything to done or in progress

221 00:22:56.350 00:22:59.859 Patrick Trainer: depending on its state. And the issue, which is really nice.

222 00:23:00.770 00:23:05.319 Ryan Luke Daque: Yeah, we don’t have to do it manually. Basically.

223 00:23:07.110 00:23:09.899 Agustin Bergoglio: yeah, that’s that’s good. Thank you.

224 00:23:18.870 00:23:21.080 Ryan Luke Daque: Yeah, I think that’s it.

225 00:23:21.120 00:23:27.250 Ryan Luke Daque: Alright, thanks. Thanks. Guys.