WEBVTT NOTE Sentence-level transcript of https://www.youtube.com/watch?v=ojJXgXLUQFc NOTE One cue per sentence. Cue ids are the line anchors on /transcripts/ojJXgXLUQFc.html. A cue ends where the next begins, or 2 s after its last word. s1 00:00:13.040 --> 00:00:14.240 Welcome everybody. s2 00:00:14.240 --> 00:00:17.119 Welcome to the AI Engineers World Fair. s3 00:00:17.119 --> 00:00:20.000 Uh thank you everyone for joining for our workshop today. s4 00:00:20.000 --> 00:00:21.199 Quick introduction for myself. s5 00:00:21.199 --> 00:00:22.160 My name is Andy Wong. s6 00:00:22.160 --> 00:00:27.279 I've been working at Alaskan for four years now and I've been working in the AI space for two and a half. s7 00:00:27.279 --> 00:00:29.679 Joining me today are two of my wonderful teammates. s8 00:00:29.679 --> 00:00:36.079 We have Shane Wolf uh manager down in Devi space as well as Andre Bohan down here as well. s9 00:00:36.079 --> 00:00:48.399 So today we're just going to go over a lot of the things how how the agentic systems are kind of changing how we are uh handling the software development life cycle and how orchestration does change everything. s10 00:00:59.760 --> 00:01:00.960 Can you repeat? s11 00:01:00.960 --> 00:01:04.400 Yeah, he's Shane just is gonna walk around give hand out numbers. s12 00:01:04.400 --> 00:01:15.280 Uh it's going to be a little bit weird at the beginning, but once we finish up the uh introduction talk, we'll be it'll make a lot of more make a lot more sense after uh after the kind of like the introduction talk kind of finalizes up. s13 00:01:15.280 --> 00:01:19.200 Um but we just want to quickly get started on the first intro slides. s14 00:01:19.200 --> 00:01:25.280 Um I do want to run through these probably quickly because since it is the AI engineers welfare we want to build. s15 00:01:25.280 --> 00:01:32.000 So I just want to give a quick rundown of everything we've been kind of building in Atlassian for the last I would say one and a half years or so. s16 00:01:32.000 --> 00:01:38.240 Uh just to give you like a highle overview of all the tools we will be offering you within Atlassian itself. s17 00:01:38.240 --> 00:01:45.520 So I think as as you guys might know AI agents have kind of changed the software development life cycle like drastically. s18 00:01:45.520 --> 00:01:53.759 Um, as time kind of goes on, you can kind of see that as we have more AI crewing tools, kind of productivity gains from those tools are kind of tapering s19 00:01:53.759 --> 00:01:56.240 um, as we have more and more tools coming out. s20 00:01:56.240 --> 00:02:03.680 And thus us at at lasting we believe that what our biggest opportunities are tapping on the on the everything left and right of code. s21 00:02:03.680 --> 00:02:11.680 So everything involving planning, orchestration, uh, the code review process as well as the operations part and such. s22 00:02:11.680 --> 00:02:19.360 We've been developing a lot of the things that um we can say like are not as apparent during the software development life cycle. s23 00:02:19.360 --> 00:02:28.640 Uh we have things such as the teamwork graph coming out as well as a bunch of different agents that we've been trying to do left or right of code um such as AI planner, AI autodev, s24 00:02:28.640 --> 00:02:34.800 reviewer, s sur um and and central of all that is something we call the AI context machine. s25 00:02:34.800 --> 00:02:42.319 Just to quickly run through all those tools, what we we what we will predominantly be testing out today is something called the Jura coding agent. s26 00:02:42.319 --> 00:02:48.640 Uh that's our that's the Atlassian coding agent natively uh that has deep organizational knowledge within the Atlastian ecosystem. s27 00:02:48.640 --> 00:02:58.400 So it's good it is a deep enterprise context agent that has uh a deep understanding of everything within the elastian ecosystem whether or not you have juror tickets confluence documents s28 00:02:58.400 --> 00:03:11.120 um it will read through email slack slack uh slack messages as well um to be able to gather all the context it needs to kind of plan out and implement a specific jur ticket that you might have s29 00:03:11.120 --> 00:03:20.159 uh within the jur coding agent itself uh you are able to uh interact with your pull request within the Jura ecosystem itself. s30 00:03:20.159 --> 00:03:26.480 So you everything is everything you'll be able to maintain and edit within the cloud. s31 00:03:28.560 --> 00:03:33.280 Addition to that you'll be a be able to assign your work to specific agents directly. s32 00:03:33.280 --> 00:03:36.319 Uh so if you don't want to use your juring agent that's fine. s33 00:03:36.319 --> 00:03:44.640 What we do provide for you as well is you are able to assign to uh agents such as claude cursor codeex whatever agent you personally prefer. s34 00:03:44.640 --> 00:03:52.239 uh you are able to assign the jur agent jur ticket to um during during that process as well. s35 00:03:52.239 --> 00:04:00.000 And finally we also have SDLC automations as you some of you might know we do have an automation process within Jura itself. s36 00:04:00.000 --> 00:04:12.159 Uh whenever there is a condition met you are able to generate like you are able to generate code uh using the jur coding agent or whatever agent you you personally decide to use. s37 00:04:18.079 --> 00:04:21.680 As we go further right through code, we do have the AI reviewer. s38 00:04:21.680 --> 00:04:26.880 Uh something similar to kind of similar to a lot of tools you've seen in the in in the industry today. s39 00:04:26.880 --> 00:04:38.800 Uh we are trying to implement uh a review of your agent to kind of go over the pull request to catch bugs, enforce standards and verify requirements every time you do make a pull request itself. s40 00:04:38.800 --> 00:04:40.639 So review does go beyond the code. s41 00:04:40.639 --> 00:04:55.440 Again, we are trying to build something that uh retrieves all the context within Jura Confluence, whatever SCM you kind of use as well as Slack to to develop like kind of like a holistic view of what your code PR needs to implement s42 00:04:55.440 --> 00:04:59.040 um and provide a review on that end. s43 00:04:59.600 --> 00:05:03.440 On top of that, we do have centralized me uh manage standards. s44 00:05:03.440 --> 00:05:10.560 Um every team, every organization kind of has their own standards that they want to implement in each uh each repository. s45 00:05:10.560 --> 00:05:25.600 Um it get it gets kind of tedious and what we're trying to do with standards is trying to trying to make it a lot easier for managers, software engineers to implement those standards in all in all the poll requests that come within a specific rep repository. s46 00:05:26.193 --> 00:05:26.720 [snorts] s47 00:05:26.720 --> 00:05:32.960 And as and additionally, this will work on all SCM platforms whether or not it's GitHub, GitLab, Bitbucket Cloud. s48 00:05:32.960 --> 00:05:38.400 um you will be able to apply standards to your your poll requests. s49 00:05:39.120 --> 00:05:47.153 One thing we are trying to do at Alaskan with our AI reviewer is we're trying to reduce the amount of like amount of noise that comes from the code review process as a whole. s50 00:05:47.153 --> 00:05:47.280 [snorts] s51 00:05:47.280 --> 00:05:59.039 Uh what we do find is that some of the code review processes uh do provide do are maybe a little bit overly nitpicky or like you know provide a little bit too many comments that are not of a high value. s52 00:05:59.039 --> 00:06:06.960 So we what we've been doing at Devi is we uh train a few models ourselves filter those and what we have done is use those models to filter out s53 00:06:06.960 --> 00:06:17.360 uh let's say low value comments or like noise that we have deemed uh and try to and filter those like unneeded comments during the poll request comments itself. s54 00:06:17.360 --> 00:06:23.600 Um as a whole we have seen that there has been a higher quality uh comment resolution as you can see the bottom here. s55 00:06:23.600 --> 00:06:37.360 uh using these models we have been able to get around 40% uh human code resolution rate which is rather close to human baseline of 45 and as always we can always measure and you know how see how impactful all these tools are. s56 00:06:37.360 --> 00:06:46.639 Um at lasting we have been building a lot of like dashboards for our various tools to see how many reviews or how many code like uh how many agents you've run s57 00:06:46.639 --> 00:06:54.080 uh to kind of give you an idea of what the value of all of our tools are uh within the devi space. s58 00:06:55.039 --> 00:06:58.960 Going on to a few like the final few few products we've been implementing recently. s59 00:06:58.960 --> 00:07:01.199 Uh something called AI autodeev. s60 00:07:01.199 --> 00:07:04.560 Um it's pretty self-explanatory what it can do. s61 00:07:04.560 --> 00:07:15.440 Uh it goes through your jur jura board uh identifies which uh which which tickets are I would say able to be able to be autonomously ran. s62 00:07:15.440 --> 00:07:21.599 Um applies all the standards initiatives um and then creates a poll request ready for review. s63 00:07:21.599 --> 00:07:25.360 Um, all of that is triggerable within the Jira ecosystem itself. s64 00:07:25.360 --> 00:07:31.840 Um, you can configure all a bunch of different needs such as maximum number of credits you want to use per month. s65 00:07:31.840 --> 00:07:37.120 Uh, maximum number of like tickets you want per per user to run autodev on. s66 00:07:37.120 --> 00:07:45.759 Um, and it's able you can also specify which repos you want to have this running on or if you want to run it on all repos within your your organization itself. s67 00:07:45.759 --> 00:07:56.639 And as always, we have a dashboard to show all the uh all the autodev runs, how it's kind of working, um how how many autodev pull requests are need to review and how many story review requirements. s68 00:07:56.639 --> 00:08:06.670 As always, we're trying to give you like as much uh metadata as possible for the engineers, for the teams to really understand like how how autodev is overall performing for you. s69 00:08:06.670 --> 00:08:08.319 [snorts] s70 00:08:08.319 --> 00:08:11.199 One other thing we do have autodev integrated on is slack. s71 00:08:11.199 --> 00:08:22.400 So once you do have uh autodev integrated within the repository itself um it will start working on a specific work items based on whether or not it's available it's able to use s72 00:08:22.400 --> 00:08:29.599 um or if there's any items that you need to uh any input to finish them more ROI insights. s73 00:08:29.599 --> 00:08:38.560 Um as always we'll we will try to measure as much as possible uh to see how what is the overall ROI you are seeing from autodev as a whole. s74 00:08:38.560 --> 00:08:48.880 And finally AI sorry uh we've all been through incidents we all understand how difficult is to diagnose and kind of go through it especially at like 2 3:00 a.m. at night. s75 00:08:48.880 --> 00:09:05.120 Um what AISR aims to do is try to grab all that context again um within previous incidents jura confluence um and try to root cause why an incident is happening and provide suggestions of how to remedi uh remediate the specific s76 00:09:05.120 --> 00:09:16.080 uh specific incidents once they do come out again it is integrated with Slack uh so you should be able to get a quick summary of what's going on uh understanding s77 00:09:16.080 --> 00:09:32.399 whether or not it's impacting multiple external customers or if it's only impacting a small section of customers and it will provide you uh detailed reason detailed uh detailed steps of how to possibly mitigate the issue that you could you are seeing s78 00:09:32.399 --> 00:09:44.320 as you can see here uh it will provide like like I said we'll provide root cause analysis um it gives you a bunch of different theories of how to possibly remediate the issue as you as you see as you see it coming s79 00:09:44.480 --> 00:09:48.800 I've been saying a lot about context and how the context will be helpful for agent itself. s80 00:09:48.800 --> 00:09:51.600 What we've been trying to build here is the AI context agent. s81 00:09:51.600 --> 00:10:02.080 Um, many of us have been working in large monor repos and all that context is scattered within the whether or not it be different different repositories itself or different pieces of documentation s82 00:10:02.080 --> 00:10:06.800 all throughout whatever uh all whatever workspace you kind of work in. s83 00:10:06.800 --> 00:10:18.560 What the AI context engine is aiming to do is try to grab all of that and create, you know, like a standardized context for you for all of your agents to kind of run in kind of like a more efficient way. s84 00:10:18.560 --> 00:10:20.720 Um, I know I've said context a lot. s85 00:10:20.720 --> 00:10:22.480 That's the biggest thing we're trying to do in the last year. s86 00:10:22.480 --> 00:10:29.839 We're trying to grab all the context as possible and try to create uh a really powerful engine for us to power our agents in the future. s87 00:10:29.839 --> 00:10:45.680 Um so as you can see here AI context engine all of our tools Jira confluence um all of our previous instance everything uh will be will be run using semantic semantic code search and we will be able to provide all that context to whatever agent you kind of use in the future. s88 00:10:46.079 --> 00:10:50.640 What we're trying to do is also make a lot of our tools like uh plug-in accessible. s89 00:10:50.640 --> 00:11:00.560 uh you so hopefully what we can do is uh allow the allow the engineer to simply just add the tool to their browser to their to their IDE or to their CLI s90 00:11:00.560 --> 00:11:04.880 and allow them to use it out of box as simply as possible. s91 00:11:05.600 --> 00:11:09.680 As I said, I wanted to keep this really short because again we're at the AI engineers world's fair. s92 00:11:09.680 --> 00:11:10.480 We want to build. s93 00:11:10.480 --> 00:11:15.920 So what we got to do here, what I do provide for you here is a small workshop for you to play around with Jurro coding agent. s94 00:11:15.920 --> 00:11:24.160 Um, you might see Shane walking around with a specific number uh specific number uh and handing handing that over to you. s95 00:11:24.160 --> 00:11:27.839 Um, that is going to be your account number. s96 00:11:27.839 --> 00:11:32.640 So, let me just quickly go over here. s97 00:11:33.440 --> 00:11:44.079 How this will work is that that number will be tied to your login and the login is the following. s98 00:12:01.440 --> 00:12:04.720 Apologies, some Wi-Fi issues. s99 00:12:09.920 --> 00:12:11.040 So, it's not just me. s100 00:12:11.040 --> 00:12:14.800 All right, got it. s101 00:12:21.040 --> 00:12:32.800 going back to this just so just so everyone does have the QR code available to them. s102 00:12:37.920 --> 00:12:39.920 Yes. s103 00:13:00.959 --> 00:13:01.200 Yeah. s104 00:13:01.200 --> 00:13:01.440 Yeah. s105 00:13:01.440 --> 00:13:01.680 Yeah. s106 00:13:01.680 --> 00:13:04.000 Let's see. s107 00:13:04.639 --> 00:13:05.120 Yep. s108 00:13:05.120 --> 00:13:07.040 Let's see. s109 00:13:07.040 --> 00:13:10.639 I will stop sharing my screen. s110 00:13:13.440 --> 00:13:13.600 Yeah. s111 00:13:13.600 --> 00:13:13.839 Yeah. s112 00:13:13.839 --> 00:13:14.079 Yeah. s113 00:13:14.079 --> 00:13:16.480 I'll do this. s114 00:13:21.519 --> 00:13:23.760 Oh god. s115 00:13:24.560 --> 00:13:26.560 Yes. s116 00:13:28.079 --> 00:13:30.320 Not quite. s117 00:13:31.839 --> 00:13:33.279 Is the internet really slow? s118 00:13:33.279 --> 00:13:35.440 Let's see. s119 00:13:36.800 --> 00:13:38.480 Oh, the Wi-Fi is quite slow. s120 00:13:38.480 --> 00:13:41.120 Did you connect to this? s121 00:13:47.040 --> 00:13:52.560 Oh, there you go. s122 00:14:09.920 --> 00:14:12.560 network changed. s123 00:14:15.519 --> 00:14:16.959 Do you still have cloud fire on? s124 00:14:16.959 --> 00:14:20.160 Try try turning it off. s125 00:14:20.480 --> 00:14:22.399 It's off. s126 00:14:22.399 --> 00:14:23.839 It is off. s127 00:14:23.839 --> 00:14:26.079 Let's see. s128 00:14:33.839 --> 00:14:36.160 How's it going? s129 00:14:36.160 --> 00:14:37.920 Yeah, we're trying to do right now. s130 00:14:37.920 --> 00:14:39.920 Yes. s131 00:14:52.320 --> 00:14:52.480 Yeah. s132 00:14:52.480 --> 00:14:55.040 Yeah, that's the Wi-Fi. s133 00:15:14.000 --> 00:15:15.279 Is that not working? s134 00:15:15.279 --> 00:15:17.920 Yeah, it's not working. s135 00:15:21.279 --> 00:15:23.680 Okay, gotcha. s136 00:15:24.720 --> 00:15:27.360 Yeah, I'm going to try mine. s137 00:15:36.880 --> 00:15:38.000 Got it. s138 00:15:38.000 --> 00:15:39.120 Yeah. s139 00:15:39.120 --> 00:15:41.900 have any questions that we can s140 00:15:41.900 --> 00:15:42.000 [snorts] s141 00:15:42.000 --> 00:15:46.079 talk through while we're figuring out Wi-Fi. s142 00:15:49.120 --> 00:15:54.880 Oh, fair, but also like um s143 00:15:54.880 --> 00:15:57.040 or stand right in front. s144 00:15:57.040 --> 00:15:58.720 Stand right here. s145 00:15:58.720 --> 00:15:59.279 Okay. s146 00:15:59.279 --> 00:16:13.839 So, while we're figuring out Wi-Fi, as I was saying, anybody have any questions that we can talk through in terms of I don't know where we're thinking of going with like the the product line, what stuff is available, what stuff people are keen on, workflows, so on. s147 00:16:13.839 --> 00:16:16.480 Uh, happy to go through any of that. s148 00:16:16.480 --> 00:16:19.120 Sorry, go for it. s149 00:16:19.920 --> 00:16:22.079 Tell us a bit more about the context engine. s150 00:16:22.079 --> 00:16:38.959 Okay, so essentially with a context engine, so um we have a number of I guess graph databases inside of Atlassian that connect to all of the data that you end up plugging into an ATL, uh into your Atlassian products as an Atlassian customer. s151 00:16:38.959 --> 00:16:46.240 So like Confluence, um Jira, Bitbucket, all that stuff is sort of connected together. s152 00:16:46.240 --> 00:16:55.279 We ma in make those connections either through inferences as you use the product or you can like have the connections be explicit. s153 00:16:55.279 --> 00:16:56.560 So like you're working on a Jurro ticket. s154 00:16:56.560 --> 00:17:00.880 The Jurro ticket is based off of like some spec document that is in Confluence. s155 00:17:00.880 --> 00:17:06.160 We also link those to like third party um um what you call a third party product. s156 00:17:06.160 --> 00:17:08.480 So like we will integrate with things like notion. s157 00:17:08.480 --> 00:17:10.400 We'll integrate with things like Slack. s158 00:17:10.400 --> 00:17:14.799 All of that stuff kind of wraps into this whole tapestry of context. s159 00:17:14.799 --> 00:17:20.959 And at the end of the day, the thing that we're uh we're building now is how do I word this? s160 00:17:20.959 --> 00:17:27.919 Uh so it's it's a gentic search on top of all of all of that context. s161 00:17:27.919 --> 00:17:40.240 And as you're this happens a lot like internally in Atlassian like we end up building on one side of the uh on one side of the ecosystem in our platform and that needs to pull in context from either s162 00:17:40.240 --> 00:17:43.840 some other team like some platform documentation. s163 00:17:43.840 --> 00:17:51.760 You have uh some other initiative that's happening on the other side of the organization that somehow happens to impact this bit of what you're building on. s164 00:17:51.760 --> 00:18:05.200 and through the context engine all of that stuff gets pulled into um your current agentic coding session like that is the stuff that Shane who's currently working on Wi-Fi so connectivity s165 00:18:05.200 --> 00:18:15.200 basically is his through line um that that's the stuff that's that he's his team's working on the things that I think we've had a lot of success with um s166 00:18:15.200 --> 00:18:30.480 has been uh so we have a new code search product that we're um we're launching um that's made it so that I think historically we've uh we've seen a lot of teams s167 00:18:30.480 --> 00:18:43.919 try to do like opt for monor repos in order to like have all their their agents like have the context available at all times that has some shortcomings and like means building out like a bunch of tooling s168 00:18:43.919 --> 00:18:54.720 uh stuff gets slow I don't know internally it's been the proc opting for a monor repo ends up meaning that you're opting for the debt of having a monor repo building the tooling around it. s169 00:18:54.720 --> 00:18:57.679 You spend a lot of time building tooling to build the thing. s170 00:18:57.679 --> 00:19:08.880 Um, and I think with a with a code context engine, we've been able to like just have product repos like monor repos because you can pull in that engine at hawk. s171 00:19:08.880 --> 00:19:18.480 It's super fast like you get code level um code level access and that means that any logic that you have that would otherwise be potentially stale, right? s172 00:19:18.480 --> 00:19:23.200 So like a conflence document is just a snapshot of a thing that happens once but like the truth is in the code. s173 00:19:23.200 --> 00:19:29.600 Being able to like access that at high throughput and like having uh good search semantics around it is been really useful. s174 00:19:29.600 --> 00:19:31.440 Sorry that was like very soap boxy. s175 00:19:31.440 --> 00:19:33.200 Does that did that answer your question? s176 00:19:33.200 --> 00:19:35.520 Okay cool. s177 00:19:38.799 --> 00:19:41.120 Anything else? s178 00:19:41.840 --> 00:19:43.840 Please. s179 00:19:47.280 --> 00:19:50.320 For example, if you have a GR s180 00:20:09.200 --> 00:20:11.039 uh sorry, so I'll I'll restate the question. s181 00:20:11.039 --> 00:20:16.000 So the first question was uh if we generate the context on the fly. s182 00:20:16.000 --> 00:20:17.280 Sorry, what was the second one? s183 00:20:17.280 --> 00:20:20.240 The second one is we get conflict information. s184 00:20:20.240 --> 00:20:21.600 Oh, how do we resolve conflicts, right? s185 00:20:21.600 --> 00:20:22.320 Because like Yes. s186 00:20:22.320 --> 00:20:23.120 Okay, that's great. s187 00:20:23.120 --> 00:20:27.200 Uh and I think there's going to be probably a bunch of talks at the conference about about memory. s188 00:20:27.200 --> 00:20:38.640 And I think uh that's kind of one of the the the hot spots of like agentic memory like you have agents that derive knowledge from like a bunch of different sources, but like at the end of the day, you need a human to tell like yes, this is actually true. s189 00:20:38.640 --> 00:20:42.480 this thing is like incidentally true or like it's true for like this specific case. s190 00:20:42.480 --> 00:20:42.880 Okay. s191 00:20:42.880 --> 00:20:54.720 Um in terms of deriving the information um it's uh usually going to be uh you don't snapshot it to some extent because of that like I think the the value of this ends up being that you have the information on the fly. s192 00:20:54.720 --> 00:20:57.840 It's available like it's the latest uh latest and greatest. s193 00:20:57.840 --> 00:21:12.240 um if you have it if you cach it then like it risks being stale and like there's all complexity around it in terms of um what the the way that we we've historically dealt with um s194 00:21:12.240 --> 00:21:24.000 authoritiveness I guess for lack of a better word um has been to provide that as uh as a feature for like the the human in the loop to like be able to say like yes cool this is verified s195 00:21:24.000 --> 00:21:30.640 this has higher authority than this other piece of content um that's harder to do with things like I don't know juror work items. s196 00:21:30.640 --> 00:21:33.520 I think juror work items in general. s197 00:21:33.520 --> 00:21:50.159 You can do it but it's just more manual work and like the risk by automating that stuff is like oh you end up like having things automated automatically end up being authoritative which is weird because you can end up in that in that same situation. s198 00:21:50.640 --> 00:21:51.600 All right. s199 00:21:51.600 --> 00:21:52.720 Are we interneted? s200 00:21:52.720 --> 00:21:53.280 I think so. s201 00:21:53.280 --> 00:21:54.720 I think we should be interneted now. s202 00:21:54.720 --> 00:21:54.960 Yeah. s203 00:21:54.960 --> 00:21:57.840 So, the p I think the page is like hopefully working for everyone now. s204 00:21:57.840 --> 00:22:03.679 Again, I'm going to put the I'm going to put the go link plus the uh QR code back up right now. s205 00:22:03.679 --> 00:22:07.360 Um, but just to clarify really quickly, everyone should have a number. s206 00:22:07.360 --> 00:22:09.280 Um, and that will be your login. s207 00:22:09.280 --> 00:22:12.640 Uh, so we will be using something called like Alaskan University. s208 00:22:12.640 --> 00:22:19.280 So, whatever number you have, it should be student and include and and then the number of whatever number that like Shane provided you earlier. s209 00:22:19.280 --> 00:22:19.490 Um, s210 00:22:19.490 --> 00:22:19.520 [snorts] s211 00:22:19.520 --> 00:22:20.640 that will be your login. s212 00:22:20.640 --> 00:22:23.760 Uh, that will be your login to test the jury coding agent. s213 00:22:23.760 --> 00:22:26.880 um and kind of just play around with it within the juror ecosystem itself. s214 00:22:26.880 --> 00:22:41.760 Uh so please go to uh so from that confluence page you should have the link to our uh to our kind of like test jur workspace um as well and from that you can kind of just like build whatever you kind of like um we do have a s215 00:22:41.760 --> 00:22:58.320 starter repository for you guys um uh it's called like Atlassian swag within bitbucket itself um I can show that in a little bit but please feel free to like once everyone logs in uh start playing around with it and start playing around with the agents directly. s216 00:23:01.919 --> 00:23:05.520 Oh, the the network is still not working. s217 00:23:08.240 --> 00:23:09.919 Oh, you are you on the same network? s218 00:23:09.919 --> 00:23:10.960 I'm on the same network. s219 00:23:10.960 --> 00:23:13.200 Yeah, it should be on the It take a little bit of time. s220 00:23:13.200 --> 00:23:15.600 It take a little bit of time to load. s221 00:23:15.600 --> 00:23:16.720 Yeah, we we Yeah. s222 00:23:16.720 --> 00:23:18.720 Yeah. s223 00:23:19.280 --> 00:23:19.520 Yeah. s224 00:23:19.520 --> 00:23:19.760 Yeah. s225 00:23:19.760 --> 00:23:21.760 Yeah. s226 00:23:22.400 --> 00:23:28.480 I think we've been we we talked with the uh the IT folks here and to see if we can uh see if there is any any resolution on that space. s227 00:23:28.480 --> 00:23:31.520 I do apologize for that. s228 00:23:36.480 --> 00:23:39.840 Yes, I can put up the QR code again. s229 00:23:40.400 --> 00:23:42.080 Yeah, no worries. s230 00:23:42.080 --> 00:23:43.520 Uh numbers, numbers, numbers. s231 00:23:43.520 --> 00:23:44.080 Yeah, sorry. s232 00:23:44.080 --> 00:23:45.120 Let me s233 00:23:45.120 --> 00:23:46.720 Did Shane go away with the numbers? s234 00:23:46.720 --> 00:23:47.520 Pardon? s235 00:23:47.520 --> 00:23:49.760 Uh where's Shane's numbers cup? s236 00:23:49.760 --> 00:23:52.320 Shane's number cup. s237 00:23:54.159 --> 00:23:56.400 Good question. s238 00:23:58.159 --> 00:24:00.400 One second. s239 00:24:02.640 --> 00:24:05.600 Need more numbers for the numbers. s240 00:24:10.159 --> 00:24:12.159 Okay. s241 00:24:18.159 --> 00:24:20.320 The best of numbers. s242 00:24:20.320 --> 00:24:22.159 I think that's a prime. s243 00:24:22.159 --> 00:24:24.159 Maybe s244 00:24:28.320 --> 00:24:30.240 I don't think anybody's able to connect to the Wi-Fi. s245 00:24:30.240 --> 00:24:30.400 Yeah. s246 00:24:30.400 --> 00:24:30.559 Yeah. s247 00:24:30.559 --> 00:24:30.880 Yeah. s248 00:24:30.880 --> 00:24:34.720 I think that was like the the the great unsolved problem of uh 2026. s249 00:24:34.720 --> 00:24:41.200 And I think if if anything's going to save us from uh AGI, it's going to be the Wi-Fi. s250 00:24:41.200 --> 00:24:45.360 Or if they fix it, then we don't need savings, so we're good. s251 00:24:47.679 --> 00:24:49.679 Uh s252 00:24:52.107 --> 00:24:52.720 [laughter] s253 00:24:52.720 --> 00:24:55.840 just you gota you got to tear the traffic, you know, you got to optimize it. s254 00:24:55.840 --> 00:24:57.279 Uh, s255 00:24:57.279 --> 00:24:57.760 okay. s256 00:24:57.760 --> 00:24:59.360 Until we figure out the Wi-Fi. s257 00:24:59.360 --> 00:25:03.600 Any other questions that I Oh, please go for it. s258 00:25:08.880 --> 00:25:10.080 Verification requirements. s259 00:25:10.080 --> 00:25:10.320 Okay. s260 00:25:10.320 --> 00:25:10.559 Sorry. s261 00:25:10.559 --> 00:25:14.080 So, the question was like, can we talk a little bit about the verification of requirements aspect? s262 00:25:14.080 --> 00:25:25.520 Um essentially that that boils down to uh when we have like a specific work item that uh the juro coding agents work agent works on that goes through and s263 00:25:25.520 --> 00:25:41.760 um like the juro coding the we expect the work item to have a checklist of hey these are the these are the acceptance criteria and as the juro coding agent uh generates the code for that particular task it'll go in and like verify s264 00:25:41.760 --> 00:25:49.840 those specific things once it thinks it's um given that it has like a the way that the jury coding agent runs is like it has a full sandbox. s265 00:25:49.840 --> 00:25:54.799 It can install things, it can run things, it can do whatever whatever it is with that specific repository. s266 00:25:54.799 --> 00:26:04.640 So as long as you have the instructions in the repository in terms of oh this is how you run this, it can pop open a browser and then like go and like run that those verification steps. s267 00:26:04.640 --> 00:26:06.720 What is the sandbox? s268 00:26:06.720 --> 00:26:08.080 Uh what is the sandbox? s269 00:26:08.080 --> 00:26:08.559 Where is it? s270 00:26:08.559 --> 00:26:10.240 Is it like hosted? s271 00:26:10.240 --> 00:26:10.720 Yeah. s272 00:26:10.720 --> 00:26:11.120 Yeah. s273 00:26:11.120 --> 00:26:12.720 Yep. s274 00:26:12.720 --> 00:26:13.600 uh s275 00:26:13.600 --> 00:26:16.640 or is it something that you can bring in? s276 00:26:16.640 --> 00:26:18.400 We don't have that ability yet. s277 00:26:18.400 --> 00:26:21.200 So it's like it's it's a it's a proprietary Atlassian sandbox. s278 00:26:21.200 --> 00:26:26.799 Uh it's again it's a container so you can mostly run like whatever it is that you want like it has Docker enabled and like the rest of stuff. s279 00:26:26.799 --> 00:26:27.039 Yeah. s280 00:26:27.039 --> 00:26:28.000 Go for it. s281 00:26:28.000 --> 00:26:32.240 Just related to billing and that type of thing your own key. s282 00:26:32.240 --> 00:26:43.600 It sounds like some if you wanted to all of these agents can use your API keys with with anthropic and open or they currently hosted models. s283 00:26:43.600 --> 00:26:52.000 Uh right now you can use uh so the billing is through Atlassian credits currently Shane am I wrong? s284 00:26:52.000 --> 00:26:55.279 Uses an API key so it uses your own API keys. s285 00:26:55.279 --> 00:26:56.000 So there's s286 00:26:56.000 --> 00:27:02.080 every one of these agents can be pointed officed wherever okay but they don't need to be s287 00:27:02.080 --> 00:27:10.720 yeah the juror coding agent supports like multiple uh I think right now we have uh all the anthropic models all the open AAI models and uh I think that's it. s288 00:27:10.720 --> 00:27:12.159 Do we have Gemini as well? s289 00:27:12.159 --> 00:27:12.960 I don't remember. s290 00:27:12.960 --> 00:27:17.520 I think we can guarantee both the cloud and the open AI models. s291 00:27:20.320 --> 00:27:20.640 Yeah. s292 00:27:20.640 --> 00:27:21.039 Yeah. s293 00:27:21.039 --> 00:27:21.279 Yeah. s294 00:27:21.279 --> 00:27:22.080 It's credits. s295 00:27:22.080 --> 00:27:23.919 It's a slightly different but like Yeah. s296 00:27:23.919 --> 00:27:24.559 Same same thing. s297 00:27:24.559 --> 00:27:27.360 It's like pay as you go. s298 00:27:29.760 --> 00:27:30.720 On top of that, s299 00:27:30.720 --> 00:27:31.360 there's usage. s300 00:27:31.360 --> 00:27:31.600 Yeah. s301 00:27:31.600 --> 00:27:32.480 Yep. s302 00:27:32.480 --> 00:27:33.840 Is this all available now? s303 00:27:33.840 --> 00:27:35.760 Sorry. s304 00:27:35.760 --> 00:27:37.039 Uh jeroding. s305 00:27:37.039 --> 00:27:37.840 Yes, it's out now. s306 00:27:37.840 --> 00:27:40.320 It's been out for for a hot minute. s307 00:27:40.320 --> 00:27:41.279 Yeah, go forward sir. s308 00:27:41.279 --> 00:28:00.240 So mention credit usage credit like for the developer like at the end of the month in terms of from a company perspective like you need to report like what kind of work that you have done or it's just s309 00:28:00.240 --> 00:28:02.240 some s310 00:28:04.000 --> 00:28:08.720 uh the if I recall correctly and I may be wrong I need I might need to follow up. s311 00:28:08.720 --> 00:28:18.320 Sorry, the question was uh how do we do uh credit reporting at the end of the month given that there's like multiple experiences that where people could you be using AI credits for right s312 00:28:18.320 --> 00:28:28.720 um if I recall correctly the way that that gets that gets split split out by product so like specific user has used a JCA with like this specific in this specific experience s313 00:28:28.720 --> 00:28:41.360 sorry JCA being juro coding agent right in this specific experience and like it uh that's kind of the the level of um uh detail you get right now not like a test based. s314 00:28:41.360 --> 00:28:41.679 Yeah. s315 00:28:41.679 --> 00:28:42.799 No, no, not yet. s316 00:28:42.799 --> 00:28:52.720 um that is something that we're working on because at the end of the day I think the thing that like will uh it's kind of becoming apparent is as uh we have like more and more people using s317 00:28:52.720 --> 00:29:07.760 uh uh agentic coding uh you want to make sure that like the ROI is there that like people when somebody is working on something is going to going to spend like I don't know uh 1 billion tokens on a thing then you want to make sure that like that is something that you actually want to have done s318 00:29:07.760 --> 00:29:13.200 as opposed to like somebody's open claw running somewhere off in the uh in the periphery. s319 00:29:18.480 --> 00:29:19.440 Uh, sorry. s320 00:29:19.440 --> 00:29:23.520 So, like what's the latest and greatest on the Wi-Fi? s321 00:29:33.760 --> 00:29:37.760 Uh, any other questions while AGI is coming? s322 00:29:37.760 --> 00:29:38.159 Yeah, sorry. s323 00:29:38.159 --> 00:29:40.320 Go ahead. s324 00:29:41.360 --> 00:29:42.799 Whoa. s325 00:29:42.799 --> 00:29:44.799 Sorry. s326 00:29:44.799 --> 00:29:46.000 Uh, login, Andy. s327 00:29:46.000 --> 00:29:46.559 There we go. s328 00:29:46.559 --> 00:29:46.799 Right. s329 00:29:46.799 --> 00:29:47.360 Right here. s330 00:29:47.360 --> 00:29:49.120 So, there should be a link. s331 00:29:49.120 --> 00:29:50.880 You can just like type this out yourself. s332 00:29:50.880 --> 00:29:53.440 Uh, 1-ish-devo.atlassian.net. s333 00:29:56.159 --> 00:30:03.600 Um, with your with your number, uh, your login will be student your number at university. s334 00:30:03.600 --> 00:30:08.000 And the password is right here, uh, ZXC ZXC on the academy. s335 00:30:08.000 --> 00:30:11.440 We're going to get it wiped at the end of the um at the end of the session. s336 00:30:11.440 --> 00:30:12.960 So like if it's open. s337 00:30:12.960 --> 00:30:13.200 Yes. s338 00:30:13.200 --> 00:30:14.080 Some questions matter. s339 00:30:14.080 --> 00:30:14.559 Doesn't matter. s340 00:30:14.559 --> 00:30:15.200 Okay. s341 00:30:15.200 --> 00:30:15.919 Thank you. s342 00:30:15.919 --> 00:30:18.320 No worries sir. s343 00:30:19.760 --> 00:30:22.080 I got you. s344 00:30:25.440 --> 00:30:26.720 So let me see really quickly. s345 00:30:26.720 --> 00:30:29.120 Let me come down. s346 00:30:33.039 --> 00:30:35.039 Yes. s347 00:30:36.480 --> 00:30:37.120 Oh god. s348 00:30:37.120 --> 00:30:40.240 Can you number you go to made with you? s349 00:30:40.240 --> 00:30:41.360 Made for you. s350 00:30:41.360 --> 00:30:41.679 Sorry. s351 00:30:41.679 --> 00:30:42.240 Made for you. s352 00:30:42.240 --> 00:30:44.080 Here. s353 00:30:44.080 --> 00:30:44.720 Made for you. s354 00:30:44.720 --> 00:30:45.840 Yes. s355 00:30:45.840 --> 00:30:47.840 There. s356 00:30:47.840 --> 00:30:50.159 Oh, there's s357 00:30:51.760 --> 00:30:52.720 this is the closest. s358 00:30:52.720 --> 00:30:54.720 Oh god. s359 00:30:54.720 --> 00:30:55.200 God. s360 00:30:55.200 --> 00:30:55.679 One second. s361 00:30:55.679 --> 00:30:56.000 Please. s362 00:30:56.000 --> 00:30:56.799 What is which? s363 00:30:56.799 --> 00:30:57.520 What is your number? s364 00:30:57.520 --> 00:30:58.640 95. s365 00:30:58.640 --> 00:31:01.039 Give it a second. s366 00:31:12.640 --> 00:31:14.640 Yes. s367 00:31:15.440 --> 00:31:17.440 Yep. s368 00:31:26.799 --> 00:31:28.000 That's a great question actually. s369 00:31:28.000 --> 00:31:32.559 So I think right now um Andre you actually I think you have a little bit more here. s370 00:31:32.559 --> 00:31:37.279 Uh Sorry, the the question was uh whether we do anything smart about model routing. s371 00:31:37.279 --> 00:31:46.720 So right off the bat with uh the geocoding agent currently I think we let you select which model like you want to route to. s372 00:31:46.720 --> 00:31:46.960 Yes. s373 00:31:46.960 --> 00:31:58.000 But I think that's the uh you can also with uh um sub agents and skills you can decide like oh this particular uh skill needs to use like so and so agent. s374 00:31:58.000 --> 00:32:00.960 I don't think we do any sorry so and so model. s375 00:32:00.960 --> 00:32:07.039 Uh we currently don't have anything out uh that would do be doing like doing smart routing. s376 00:32:07.039 --> 00:32:14.480 That is something that we're working on with uh in in autodev to help you essentially figure out like oh this specific task this is a high cool level. s377 00:32:14.480 --> 00:32:22.320 We've seen success with it and like give you the metrics in order to be able to like have that decision and make sure that that's a policy that you want to enforce or not. s378 00:32:22.320 --> 00:32:29.600 Please stacking that in terms of like you've got your agents then you've got your skills. s379 00:32:29.600 --> 00:32:36.960 Do we define the agent or agents preset by you and then we define the skills that they might automot? s380 00:32:44.720 --> 00:32:56.640 Uh the skills are basically up to you like we have like a bunch of bunch of skills but specifically for the juro coding agent there's a um bifurcation to some extent between like s381 00:32:56.640 --> 00:32:59.440 rode dev s382 00:32:59.440 --> 00:33:17.360 uh as as they were called previously um and with the juro coding agent that's essentially we uh give you a harness and there's model selection at the top level of the harness and then with the skills you can figure out like right all right like this is for this specific task, for this specific skill, I want this s383 00:33:17.360 --> 00:33:21.360 model, this sub agent needs to be using blah blah blah. s384 00:33:25.120 --> 00:33:27.120 Um, s385 00:33:28.559 --> 00:33:28.880 yeah. s386 00:33:28.880 --> 00:33:29.120 Yeah. s387 00:33:29.120 --> 00:33:32.320 And that is up to up to the model. s388 00:33:32.320 --> 00:33:34.320 Yeah. s389 00:33:43.039 --> 00:33:45.760 Oh, we're still doing Wi-Fi. s390 00:33:45.840 --> 00:33:47.039 Oh. s391 00:33:47.039 --> 00:33:48.159 Uh, okay. s392 00:33:48.159 --> 00:33:51.519 Well, we're still doing Wi-Fi, please. s393 00:33:56.399 --> 00:33:56.720 Yeah. s394 00:33:56.720 --> 00:33:58.640 Yeah, that's probably what we're going to do. s395 00:33:58.640 --> 00:33:59.200 Uh, sorry. s396 00:33:59.200 --> 00:34:05.440 The question was like, can we keep the demo site open for a hot minute so that like when people actually get to Wi-Fi, they can actually go through it? s397 00:34:05.440 --> 00:34:07.919 Yes, I think that's a that's a reason to ask. s398 00:34:07.919 --> 00:34:10.079 Thank you. s399 00:34:10.079 --> 00:34:10.720 Um, s400 00:34:10.720 --> 00:34:13.520 how long will this be? s401 00:34:13.839 --> 00:34:16.000 We can keep it open for as long as we really want. s402 00:34:16.000 --> 00:34:20.960 I think maybe for maybe if like for the duration of AI AIS fair, s403 00:34:20.960 --> 00:34:30.399 uh, we'll keep the we'll keep the workshop open and so when we do have Wi-Fi, I do apologize for the Wi-Fi issues, you can kind of like play around with for yourself and kind of like play around with the different tools we we will we do have. s404 00:34:30.399 --> 00:34:39.040 Uh, one thing to sort of keep in mind as you're like inputting stuff onto these sites, as you can tell, the login are like very secure. s405 00:34:39.040 --> 00:34:47.440 So, like don't put in anything that you wouldn't be comfortable putting on a GitHub gist somewhere or like on a a bit bucket snippet. s406 00:34:47.440 --> 00:34:48.639 So, yeah. s407 00:34:48.639 --> 00:34:55.359 Uh, keep that in mind as you're kind of adding stuff to and and messing around with all these things. s408 00:34:55.359 --> 00:34:57.359 Um, all right. s409 00:34:57.359 --> 00:35:00.560 Any other questions that we can go into? s410 00:35:00.560 --> 00:35:02.800 Please, sir. s411 00:35:05.839 --> 00:35:09.839 What's the importance of cleaning up stale context? s412 00:35:10.160 --> 00:35:10.560 Interesting. s413 00:35:10.560 --> 00:35:11.119 Can you say more? s414 00:35:11.119 --> 00:35:13.920 Like what do you what do you mean? s415 00:35:14.800 --> 00:35:17.440 Atlas for 15 20 years. s416 00:35:17.440 --> 00:35:20.400 There's tons of product s417 00:35:23.200 --> 00:35:25.839 out of all the things. s418 00:35:26.079 --> 00:35:26.880 It's a good Okay. s419 00:35:26.880 --> 00:35:27.119 Sorry. s420 00:35:27.119 --> 00:35:28.560 So like the good question. s421 00:35:28.560 --> 00:35:34.560 Um the the the question is essentially so like you have a large organization that's been using Atlassian for like 10 15 years. s422 00:35:34.560 --> 00:35:45.760 Um how important is it to like prune some of that stuff so like it doesn't end up showing in search results or like the agent doesn't end up picking it up while it's looking for random stuff to give you an answer to the question. s423 00:35:45.760 --> 00:35:47.114 uh s424 00:35:47.114 --> 00:35:47.839 [sighs] s425 00:35:47.839 --> 00:35:56.560 uh our experience has mostly been okay with this by in in large part because we tend to favor like more recent results. s426 00:35:56.560 --> 00:36:06.000 Uh I think that's where authoritiveness sort of comes in where like if you have something that's been around for 10 years and it's still like the whatever tip of the spear is like that document for 10 years ago. s427 00:36:06.000 --> 00:36:19.760 um those I think we currently shipped a thing in uh well shipped a thing a while back in in um conference that's the verified badge like that's basically the thing that we use internally to say like yes cool this is like still the latest and greatest and that this context s428 00:36:19.760 --> 00:36:43.599 should be favored uh across others Hello. s429 00:36:43.599 --> 00:36:44.800 Okay, awesome. s430 00:36:44.800 --> 00:36:47.119 I think we've had a few questions about standards. s431 00:36:47.119 --> 00:36:49.839 Uh, unfortunately, the standards aren't built into the demo itself. s432 00:36:49.839 --> 00:36:55.359 Uh, but I do want to take a little bit of time just to like show you how standards kind of work within, uh, the Robo Dev product itself. s433 00:36:55.359 --> 00:37:02.000 uh and kind of kind of like how we would be assigning standards overall uh to your specific repositories itself. s434 00:37:02.000 --> 00:37:03.599 Um so just a quick demo here. s435 00:37:03.599 --> 00:37:05.280 Uh so this is the standards page. s436 00:37:05.280 --> 00:37:10.320 Uh we have a few built in already, but what I want to show you right now is like you know how to create your own standards. s437 00:37:10.320 --> 00:37:13.520 Um there's also a bunch of different templates you are able to select. s438 00:37:13.520 --> 00:37:16.480 Uh but what I'm going to do is just create a new one. s439 00:37:16.480 --> 00:37:18.400 Uh new standard. s440 00:37:18.400 --> 00:37:22.240 Uh give a quick description of how what the standards really is. s441 00:37:22.240 --> 00:37:26.480 Uh so so this one would be design token enforcement. s442 00:37:26.480 --> 00:37:35.280 So if we go to standard standard name go to call it design token enforcement uh add the description itself and add the rules. s443 00:37:35.280 --> 00:37:39.040 So after you add the standard s444 00:37:39.120 --> 00:37:39.760 [snorts] s445 00:37:39.760 --> 00:37:49.040 and the rules itself you can specify which repositories you want these standards to be applied to. s446 00:37:49.040 --> 00:37:53.119 uh you can select manually all the repositories or repositories matching a pattern. s447 00:37:53.119 --> 00:38:02.240 Um so as these repositories load you can decide okay let's apply to these following repositories and then after every all that is set up you can just create your standard s448 00:38:02.240 --> 00:38:19.760 um like I said within software we have our own standards um instead of trying to enforce it uh manually trying to like enforce that within the PR process itself what we can do is you can set up the standard set up set up which which repositories you want the standard to uh be enforced on and the robo s449 00:38:19.760 --> 00:38:30.160 dev jur code like the coding agent itself will be running through the pull request process and adding those uh adding comments to see if those p if your pull request isn't adhering to all the standards that s450 00:38:30.160 --> 00:38:33.440 um are kind of the specified within the standard itself. s451 00:38:33.440 --> 00:38:41.520 Um a few other things I did want to demo uh something called AI planner as well as like autodev. s452 00:38:41.520 --> 00:38:51.280 Um we can't give a full-fledged demo but we do have what we do want to iterate that like you know we're trying to go from the left or right of code and I would say like AI planner and autodev are more on the left of code. s453 00:38:51.280 --> 00:39:00.960 So, um I don't know if we have a full demo, but I can kind of talk about it with the uh slideshow itself. s454 00:39:08.400 --> 00:39:10.240 So, we're going to get start uh autodev. s455 00:39:10.240 --> 00:39:12.560 Actually, Andre, uh you've been working on autodev pretty heavily. s456 00:39:12.560 --> 00:39:18.480 Can you kind of like give us a brief uh brief overview of what the main goals you want to accomplish with AutoDev? s457 00:39:18.480 --> 00:39:23.200 Um, and kind of kind of works as he sets up his mic. s458 00:39:23.760 --> 00:39:26.960 Trying to figure out how mics work. s459 00:39:26.960 --> 00:39:29.040 Supposed to touch it. s460 00:39:29.040 --> 00:39:30.000 Not supposed to touch it. s461 00:39:30.000 --> 00:39:30.560 Oh, there we go. s462 00:39:30.560 --> 00:39:31.440 Oh my god. s463 00:39:31.440 --> 00:39:32.400 Thank you. s464 00:39:32.400 --> 00:39:34.320 Um, all right. s465 00:39:34.320 --> 00:39:34.560 Sorry. s466 00:39:34.560 --> 00:39:37.760 Let me I'll just be here then. s467 00:39:37.760 --> 00:39:38.720 Um, okay. s468 00:39:38.720 --> 00:39:45.200 So, for for autodev, the this is essentially our for into like the AISDLC orchestration space. s469 00:39:45.200 --> 00:39:58.000 Um the thought process is to basically get you back some capacity from all the in order to let your engineering team focus on feature work like the uh AI the for autodev we s470 00:39:58.000 --> 00:40:11.680 the initial um the initial release is going to focus on grabbing work items from your backlog and grabbing the ones that we feel confident that uh the agents can handle autonomously s471 00:40:11.680 --> 00:40:25.599 and then just putting up PRs for your team to review in order to make sure that like we're working on the right things that we're abiding by the right standards and then sort of iterating on your harness as well so that we can get code bases agent ready in a fairly light touch way. s472 00:40:25.599 --> 00:40:34.960 Um do you want to oh sorry do you want okay and this this essentially kind of goes where uh uh establishes where where we're trying to go. s473 00:40:34.960 --> 00:40:45.440 So um you we right now again the definition is going to be like we will we will grab your backlog and then we will allow you to also help us s474 00:40:45.440 --> 00:40:50.400 identify which specific things you think auto you would like autodev to to grab. s475 00:40:50.400 --> 00:40:55.839 Um we have some some thoughts around like where we would like to this to go next. s476 00:40:55.839 --> 00:41:02.240 So there's standards enforcement um and I think Andyy's going to talk probably talk about that in the context of the review. s477 00:41:02.240 --> 00:41:12.400 So the way that we think about that is I don't know standards are to some extent pretty similar to linting to some degree but they're a lot more uh s478 00:41:12.400 --> 00:41:16.960 loose uh when you have the uh agents enforce them. s479 00:41:16.960 --> 00:41:31.760 We have one part of the I guess the the the linking workflow in forward-looking things in the reviewer right like you add standards to code reviewer and that takes care of any code that will end up landing in your codebase going forward. s480 00:41:31.760 --> 00:41:44.560 uh we have a workflow for autodev where we would be able to scan existing part like your existing codebase and then figure out how we land uh PRs progressively in order to get all of your existing stuff into compliance. s481 00:41:44.560 --> 00:41:55.040 There's some complexity there in terms of like so nobody's going to approve a PR that's like change half your codebase because now you have a new standards uh or nobody's going to approve anything that like causes like a high amount of churn. s482 00:41:55.040 --> 00:41:59.599 So with autodev we want to be mindful of how much work we're actually hoisting on your team. s483 00:41:59.599 --> 00:42:03.920 uh nobody wants to spend time reviewing uh 50 PRs a day. s484 00:42:03.920 --> 00:42:20.079 So like we have configuration for letting you say cool like my team wants to get assigned I don't know each individual gets assigned three five PRs a week and like those are the things so that makes helps you get like incremental progress and also frees up your team to actually do the feature work that s485 00:42:20.079 --> 00:42:25.359 they uh they actually want to do and like may move forward your u other initiatives. s486 00:42:25.359 --> 00:42:28.880 Um and then yeah that kind of goes into like both initiatives as well. s487 00:42:28.880 --> 00:42:33.599 I think the the space there is I kind of think about it as organizational steering. s488 00:42:33.599 --> 00:42:46.720 So this happens a lot in uh bigger organizations where you have I don't know everybody's trying to move from whatever some version of Rails to a different version of Rails and it's a big thing and we have like 10,000 services and like those are pretty similar s489 00:42:46.720 --> 00:42:53.520 uh and some of the learnings some of the memory some of the workflows that you go through are fairly uh sy. s490 00:42:53.520 --> 00:43:03.119 So you would want something that is concentrated on one team and they're just like spinning up PRs and like the agents essentially do the do the work and then you'd only have a human do the coordination s491 00:43:03.119 --> 00:43:07.359 and then we also tie the loop around that with uh some sort of memory. s492 00:43:07.359 --> 00:43:07.680 Sorry. s493 00:43:07.680 --> 00:43:14.960 So again that's that's sort of the vision that that's where we see this going and that's where like we see this being beneficial to our workflows because large organization s494 00:43:14.960 --> 00:43:25.040 bunch of services that's where we spend a bunch of time that like we would like to not have engineers be involved in that deeply and we see agents as like a really big opportunity to to move that ball forward. s495 00:43:25.040 --> 00:43:41.440 I think you can grab on to okay um this is the other thing uh if uh autodev is doing work uh on your team's behalf autonomously you kind of make want to make sure that like you are aware of like what it's doing and want to be are able to interject s496 00:43:41.440 --> 00:43:55.119 when you want to when it's either picking up the wrong thing or like it's not assigning the wrong the right person and we will learn from that experience and uh the reason why you want that to be in your jura board is because that's the thing that most people stare at like it or not when they're trying s497 00:43:55.119 --> 00:43:58.319 to like figure out what's the next thing to to work on. s498 00:43:58.319 --> 00:44:14.480 Um, and we were talking about this before uh just the um the the work in progress limit is the thing that we were that I was mentioning where you want to make sure that like your team isn't getting inundated with PRs or like other work that might not be necessarily the work that you want to drive forward. s499 00:44:14.480 --> 00:44:24.560 Um the other thing that we we're also thinking through is um limiting cost which I think is very much uh in the site guys now. s500 00:44:24.560 --> 00:44:26.560 Like you don't really want agents running roughshot. s501 00:44:26.560 --> 00:44:36.079 Like if you want to spin up a a stream of work for autodev, you want to be able to set limits for how much we want to spend per ticket, how much we want to spend per week. s502 00:44:36.079 --> 00:44:41.119 All that stuff seems like it's uh just table stakes at this point. s503 00:44:41.119 --> 00:44:52.720 Um yeah this is essent like the whenever we whenever our dev starts working on stuff we will um give you like this this broad reviewer like all right how much how many things needs s504 00:44:52.720 --> 00:45:07.280 uh oh man I I forgot to mention this okay need requirements ah uh for whenever we troll through backlogs there's um a very high likelihood that we have work items that are in your backlog that don't have that are underspecified s505 00:45:07.280 --> 00:45:12.160 like there's a thing that somebody put in like that has just one summary line that says do the thing. s506 00:45:12.160 --> 00:45:21.280 Um those can be valuable but we don't know and we don't have the confidence that we can work on them because we don't know exactly what the goals of them are. s507 00:45:21.280 --> 00:45:26.079 And that's where like we will pull a pull in a human in the loop and say hey we have this work item. s508 00:45:26.079 --> 00:45:27.839 What is what actually is this thing? s509 00:45:27.839 --> 00:45:29.119 Do we want it to work on it? s510 00:45:29.119 --> 00:45:30.640 Then can you help me specify it? s511 00:45:30.640 --> 00:45:35.520 And then once we get to a large good level of specification that's when like we can get to actually working on it. s512 00:45:35.520 --> 00:45:39.920 And that's where again needs review is when we pull in a human to to actually do the review. s513 00:45:39.920 --> 00:45:49.359 Um the other bit that we've also noticed with some of these autonomous workflows is the the the reviewing step is where things go to die. s514 00:45:49.359 --> 00:45:53.680 Um because people again are uh busy with other things. s515 00:45:53.680 --> 00:46:01.920 Um, and that's one of the systems that we're putting together in order to reach out to people at the opportune time in order to pull them in and go like, "Hey, can you like s516 00:46:01.920 --> 00:46:08.240 take an eye to take a look at this real quick and like figure out whether um whether it's good to merge or like it's it needs to go away." s517 00:46:08.240 --> 00:46:15.520 And the [clears throat] the chief vehicle for that is going to be Slack for for the the initial release of AutoDev. s518 00:46:15.520 --> 00:46:28.400 Um, and then again, we were talking about this, the ROI insights, you making sure that like you're getting um, you're not overpaying for the amount of work that we're doing and then it's work that you actually want to get done and being able to like sort of push the limits of like how much stuff you actually want s519 00:46:28.400 --> 00:46:30.560 us to look at. s520 00:46:30.560 --> 00:46:31.359 Awesome. s521 00:46:31.359 --> 00:46:39.760 So, I think the one I think the one part that we want to touch upon again uh, really quickly is this part, the the planning part. s522 00:46:39.760 --> 00:46:42.720 Uh, so we do have one more product out coming out called AI Planner. s523 00:46:42.720 --> 00:46:54.480 Uh I would say this is more geared for I would say like um say uh if you have a spike or project managers or say you're a manager that wants to see if something is feasible for your current repository. s524 00:46:54.480 --> 00:46:55.920 That's where AI planner comes in. s525 00:46:55.920 --> 00:47:06.000 you can kind of provide your idea of what you want to build and what the a planner will kind of do again goes through our context engine kind of figures out okay what is in your repository s526 00:47:06.000 --> 00:47:13.839 what is your plan and is it feasible and if it is feasible it will tell you and the second part is it will kind of give you a breakdown of what you need to implement s527 00:47:13.839 --> 00:47:22.960 um if you do want to build out the spec specific uh specific feature or the specific you know idea that you do kind of have it will give you the plan uh and it kind of give you s528 00:47:22.960 --> 00:47:28.319 a few ways to execute it um talked a lot about the context engine. s529 00:47:28.319 --> 00:47:29.680 Um Shane is actually here. s530 00:47:29.680 --> 00:47:31.280 He's working a lot on code chat. s531 00:47:31.280 --> 00:47:38.480 Um Shane, do you want to talk a little bit more about the code chat capabilities and like all everything within like a context engine itself? s532 00:47:39.680 --> 00:47:41.072 Are there slides? s533 00:47:41.072 --> 00:47:41.599 [snorts] s534 00:47:41.599 --> 00:47:42.560 Are there slides? s535 00:47:42.560 --> 00:47:43.520 Uh there's no slides. s536 00:47:43.520 --> 00:47:44.160 All right. s537 00:47:44.160 --> 00:47:47.680 Um yeah, so he was mentioning uh context engine code chat. s538 00:47:47.680 --> 00:47:53.440 The way we've been thinking about it and at at lastian we have many repositories. s539 00:47:53.440 --> 00:48:05.599 is I think we have 15,000 repositories across the company and asking a simple question like where do I implement this change or where is X Y and Z implemented or what services are using this other service s540 00:48:05.599 --> 00:48:17.119 are really really complicated and so at the core of the context engine it is narrowing down the retrieval and the search space to scope it to like where this answer is most likely s541 00:48:17.119 --> 00:48:26.079 uh lives within the code and then reaching out from there and like digging into the code to answer the questions and then on top of that len and all of the s542 00:48:26.079 --> 00:48:27.920 um information from the teamwork graph. s543 00:48:27.920 --> 00:48:33.040 So documentation historically out ofdate and inaccurate. s544 00:48:33.040 --> 00:48:38.559 If you're searching only across documentation, you're going to get something that was true at some point in time. s545 00:48:38.559 --> 00:48:40.960 Um but it's also a lot of really valuable information. s546 00:48:40.960 --> 00:48:45.760 So this project built this thing at this point in time. s547 00:48:45.760 --> 00:48:46.559 It was true. s548 00:48:46.559 --> 00:48:47.599 It's no longer true. s549 00:48:47.599 --> 00:48:54.079 But the reason why those things are decided at the time are valuable for making changes in the future. s550 00:48:54.079 --> 00:49:04.240 And so we can leverage the code as it exists today, the decisions that went into the code as it existed at some point in the past and use that to inform s551 00:49:04.240 --> 00:49:11.280 um implementation paths or or or planning for you know why we should change code in a certain way. s552 00:49:11.280 --> 00:49:12.480 Does that make sense? s553 00:49:12.480 --> 00:49:13.680 Any questions? s554 00:49:13.680 --> 00:49:27.119 But yeah, like historical information about why decisions were made are something that's often left out of um agent generated code, but it's really important um historical context for like not re remaking mistakes of the past. s555 00:49:27.119 --> 00:49:30.240 And so that's how we think about the context engine. s556 00:49:30.240 --> 00:49:30.640 Awesome. s557 00:49:30.640 --> 00:49:31.200 Thanks. s558 00:49:31.200 --> 00:49:33.839 We have a question over here. s559 00:49:39.200 --> 00:49:53.440 Yeah, it's um a lot of it is used like AI planner is leveraging it for a lot of things that I don't know if either of you can speak more about but a AI planner is using it a lot right so if you have something that you're trying to figure out s560 00:49:53.440 --> 00:50:02.319 um how to build it can look at documents for projects related to it context around it and so it's it's kind of at the core of a lot of these new capabilities we're building out s561 00:50:02.319 --> 00:50:06.800 um autodev uses it um and you could probably speak more to it. s562 00:50:06.800 --> 00:50:12.319 Um AI reviewer doesn't use it a lot today, but we're starting to experiment with how to pull more um into there. s563 00:50:12.319 --> 00:50:16.640 And then the other place it's exposed today is within Robo Chat. s564 00:50:16.640 --> 00:50:22.319 So if you're talking to Robo and um we're in early access with a few customers now. s565 00:50:22.319 --> 00:50:33.680 just announced it a couple months ago, but if you ask it code related questions today, it answers those through um mostly Confluent Search, some other uh connected tooling. s566 00:50:33.680 --> 00:50:38.640 And then if you have GitHub or Bitbucket, it will use their search APIs. s567 00:50:38.640 --> 00:50:55.440 Um but if you are you know have this new capability called code context enabled it will have like a more cohesive understanding of the code grounded in the truth today the historical decisions that went into it and answer those types of code related questions more effectively s568 00:50:57.200 --> 00:51:02.319 sorry that access s569 00:51:03.040 --> 00:51:14.720 yeah so the teamwork graph CLI um if you've seen that that's where we're starting to expose a lot of these capabilities Um the context engine itself is more agentic search is the way to think about it. s570 00:51:14.720 --> 00:51:24.880 We have a lot of the core primitives that we use uh that we'll expose to your agents through the CLI but I think over time you know we might consider moving more of those into there as well. s571 00:51:24.880 --> 00:51:30.800 Yeah, team graph CLI has access to all the um code context skills. s572 00:51:31.599 --> 00:51:33.119 Question in the back. s573 00:51:33.119 --> 00:51:34.559 Yeah, I had two questions. s574 00:51:34.559 --> 00:51:36.400 The first was on standards. s575 00:51:36.400 --> 00:51:39.440 Um it looked like standards were like pros and marked down. s576 00:51:39.440 --> 00:51:40.240 Is that correct? s577 00:51:40.240 --> 00:51:40.800 That's correct. s578 00:51:40.800 --> 00:51:41.440 Yes. s579 00:51:41.440 --> 00:51:47.359 Do you have any way to map those to something deterministic like a CI check? s580 00:51:48.079 --> 00:51:53.359 So kind of like kind of like what we say like a check like a checkbox kind of like uh scenario, right? s581 00:51:53.359 --> 00:52:01.920 Um I think just off the top of my head what we would probably have to do is kind of like implement that within a the markdown itself. s582 00:52:01.920 --> 00:52:10.000 But that's actually really interesting use case where you can just provide it like specific like can like binary yes does this satisfy no does this not satisfy. s583 00:52:10.000 --> 00:52:18.240 Um so I currently I don't think so I don't think we can prov we provides that out of box but I will relay that over to the standards team to see if we can get that in the future. s584 00:52:18.240 --> 00:52:21.280 Yeah, that could be really neat to make it a little bit more. s585 00:52:21.280 --> 00:52:23.359 That would be awesome. s586 00:52:23.359 --> 00:52:25.359 Right. s587 00:52:27.119 --> 00:52:33.119 Epic that you can break it down and decompose it into multiple ages like waves of ages. s588 00:52:33.119 --> 00:52:39.359 Usually there's foundational work that will work down the line. s589 00:52:39.599 --> 00:52:44.400 foundational unblock. s590 00:52:49.760 --> 00:52:50.000 Right. s591 00:52:50.000 --> 00:53:03.520 Just to reiterate uh I think your use case is kind of similar like where you have an epic um and you maybe want that epic to be like broken down into like smaller issues and then within those smaller issues you can have like you know like a wave of agents kind of like execute each smaller piece. s592 00:53:03.520 --> 00:53:04.079 Is that correct? s593 00:53:04.079 --> 00:53:11.440 Would you want like maybe like one step further once you implement like that first wave like is it if like if we could get a second wave going? s594 00:53:11.440 --> 00:53:11.920 Yes. s595 00:53:11.920 --> 00:53:12.319 Okay. s596 00:53:12.319 --> 00:53:12.640 Awesome. s597 00:53:12.640 --> 00:53:12.880 Yeah. s598 00:53:12.880 --> 00:53:13.040 Yeah. s599 00:53:13.040 --> 00:53:14.640 Good great question. s600 00:53:14.640 --> 00:53:16.640 So I actually played around with this yesterday. s601 00:53:16.640 --> 00:53:24.319 Um with JCA what you can do you can take an epic uh you can get JCA can you break this down into like smaller issues for me and then s602 00:53:24.319 --> 00:53:29.760 uh it will be able it it is able to break it into like smaller juror task if you provided a jur epic. s603 00:53:29.760 --> 00:53:43.520 After that what you can what you are able to do um there's a few like automation rules that you can kind of do let's say let's say like uh with an automation let's say like you have the J epic kind of like be broken down once it's broken down you can have automations kind of like re start s604 00:53:43.520 --> 00:53:49.440 a JCA uh instance has start a JCI kind of session and it is able to like autonomously kind of like develop each s605 00:53:49.440 --> 00:54:03.839 and I don't think there's anything that would prevent you from kind of extrapolating that further right like here's all the stories I want break these down into multiple epics right the foundational work the like V 0, V1, V2, and then just have it, you know, further refine the tickets into there. s606 00:54:03.839 --> 00:54:06.960 And so I think nothing would prevent you from doing s607 00:54:06.960 --> 00:54:07.920 Oh, you can keep going. s608 00:54:07.920 --> 00:54:09.520 Yeah, you can do flexible. s609 00:54:09.520 --> 00:54:11.760 Keep going. s610 00:54:16.000 --> 00:54:23.520 Sounds more like a JCA automation kind of like I'm not sure if Robo directly could do it, but I definitely think you can do with automations. s611 00:54:23.520 --> 00:54:23.760 Yeah. s612 00:54:23.760 --> 00:54:27.359 And I think autodev might be able to do some of that. s613 00:54:27.359 --> 00:54:27.839 Auto dev. s614 00:54:27.839 --> 00:54:28.240 Yes. s615 00:54:28.240 --> 00:54:28.720 Okay. s616 00:54:28.720 --> 00:54:30.452 I do that with gas. s617 00:54:30.452 --> 00:54:30.960 [laughter] s618 00:54:30.960 --> 00:54:32.000 You guys are doing great. s619 00:54:32.000 --> 00:54:32.400 Thank you. s620 00:54:32.400 --> 00:54:34.400 Yeah, no worries. s621 00:54:34.400 --> 00:54:36.079 question. s622 00:54:36.079 --> 00:54:50.720 I just want to understand that actually my code is I change the code. s623 00:54:58.079 --> 00:55:04.319 So just to reiterate uh let's say you make a code change within like your IDE. s624 00:55:04.319 --> 00:55:05.040 Oh yeah. s625 00:55:05.040 --> 00:55:11.359 Um so I think you have to think about with JCA that like it it is isolated very much the agent sessions. s626 00:55:11.359 --> 00:55:16.000 So if you say like make a change within the agent session itself um then it will be real time. s627 00:55:16.000 --> 00:55:25.040 Uh if you make it make a change within the VS code instance within the cloud then yes all those changes will happen real time and you will be able to be able to reflect that within the cloud session itself. s628 00:55:25.040 --> 00:55:29.280 Um whatever changes you make on the ID it's a separate kind of like a separate instance right. s629 00:55:29.280 --> 00:55:41.760 Uh so yeah with with agents think about it like all within the cloud itself any change you make in Jira session jur agent session itself within jura then then then all of that will be real time s630 00:55:42.480 --> 00:55:45.040 question over here s631 00:55:54.000 --> 00:56:01.680 you I guess you s632 00:56:03.599 --> 00:56:11.760 so just to iterate like so it sounds like your instance is more like on prem and you want to run a bunch of like the agents like on on prem. s633 00:56:11.760 --> 00:56:12.720 Is that correct? s634 00:56:12.720 --> 00:56:28.559 Not even like I guess like how is the develop development environment like is there like a docker file that we would have to use to s635 00:56:30.720 --> 00:56:32.400 that's s636 00:56:32.400 --> 00:56:32.960 that's docker. s637 00:56:32.960 --> 00:56:33.599 It is docker. s638 00:56:33.599 --> 00:56:33.760 Yeah. s639 00:56:33.760 --> 00:56:34.400 Yeah. s640 00:56:34.400 --> 00:56:36.319 Okay. s641 00:56:36.319 --> 00:56:36.559 Cool. s642 00:56:36.559 --> 00:56:40.480 I think we have time for one more question. s643 00:56:44.480 --> 00:56:45.359 That's a great question. s644 00:56:45.359 --> 00:56:51.040 Uh, we actually just started uh supporting skills I think just not too long ago. s645 00:57:02.400 --> 00:57:02.720 Right. s646 00:57:02.720 --> 00:57:03.280 Yes. s647 00:57:03.280 --> 00:57:10.720 So yeah, I think for just AI reviewer itself, we do support skills like uh like I think less than a month ago, we released leads. s648 00:57:10.720 --> 00:57:13.119 Sounds good. s649 00:57:13.839 --> 00:57:14.240 Cool. s650 00:57:14.240 --> 00:57:15.200 All right, thanks everyone. s651 00:57:15.200 --> 00:57:15.920 We got to go. s652 00:57:15.920 --> 00:57:17.680 We'll be around if there's any other questions. s653 00:57:17.680 --> 00:57:18.240 But s654 00:57:18.240 --> 00:57:18.960 yeah, thanks a lot. s655 00:57:18.960 --> 00:57:21.280 Thanks a lot everyone for joining uh joining our workshop. s656 00:57:21.280 --> 00:57:24.799 Thanks a bunch for the uh the tech staff for getting all of our Wi-Fi issues sorted out. s657 00:57:24.799 --> 00:57:25.839 Big shout out to them. s658 00:57:25.839 --> 00:57:29.359 Um hope you guys have a great rest of your week during the uh during the conference. s659 00:57:29.359 --> 00:57:32.000 Thank you so much, guys. s660 00:57:45.644 --> 00:57:47.644 [music]