1 00:00:00,312 --> 00:00:02,574 code reviews can be a pain, right? 2 00:00:02,855 --> 00:00:07,780 A lot of people don't like them, and we see many cases where code reviews are not that great. 3 00:00:07,800 --> 00:00:10,824 And again, surprise, AI can help. 4 00:00:10,824 --> 00:00:17,451 We see a rise in AI code reviewers that can help you assess your code or improve it before you reach production. 5 00:00:17,451 --> 00:00:18,607 Let's have a look. 6 00:00:25,625 --> 00:00:29,447 Hi everyone, welcome to technology explorations at Dataminded. 7 00:00:29,447 --> 00:00:33,140 In this series, we give you an initial look in new or interesting technologies. 8 00:00:33,140 --> 00:00:35,981 My name is Jonny, knowledge lead here at Dataminded. 9 00:00:35,981 --> 00:00:39,634 And today we're going to look into AI assistance for code reviews. 10 00:00:39,634 --> 00:00:41,757 And for that, I brought with me Hannes. 11 00:00:41,757 --> 00:00:42,618 Welcome Hannes. 12 00:00:42,618 --> 00:00:43,887 Hi, thanks for having me. 13 00:00:43,887 --> 00:00:44,684 How are you doing today? 14 00:00:44,684 --> 00:00:45,250 All good? 15 00:00:45,250 --> 00:00:46,377 All good, all good. 16 00:00:46,377 --> 00:00:47,997 Ready to show you some things. 17 00:00:47,997 --> 00:00:50,639 So what brought you to this topic? 18 00:00:50,639 --> 00:00:52,380 AI assisted code reviews. 19 00:00:52,380 --> 00:00:54,092 What is the reason this triggered you? 20 00:00:54,092 --> 00:01:01,346 Yes, so I think we were all quite amazed when we first used whatever model it was. 21 00:01:01,346 --> 00:01:04,629 Was it a GPT model or Gemini model or Claude or whatever. 22 00:01:04,629 --> 00:01:05,700 or a lot of people are... 23 00:01:05,700 --> 00:01:09,863 utilizing that or at least to some extent should, I think, that's my opinion. 24 00:01:09,863 --> 00:01:14,937 But you've also noticed, think, Jonny, that it sometimes doesn't know really what it's doing. 25 00:01:14,937 --> 00:01:22,022 It doesn't have enough context, which is most of our time our fault, but sometimes it also just makes mistakes and you've been there. 26 00:01:22,022 --> 00:01:24,484 It's not always a walk in the park. 27 00:01:24,484 --> 00:01:25,967 if you accept that and 28 00:01:25,967 --> 00:01:28,389 want to go to production, you go through PRs. 29 00:01:28,389 --> 00:01:35,198 But after a while, the PRs start to become so big because people are producing so much code because it's so easy to produce code. 30 00:01:35,198 --> 00:01:40,515 And I was really interested, if you're automating software engineering itself, writing the code, can we also automate 31 00:01:40,515 --> 00:01:42,989 or leverage AI to do the PRs itself. 32 00:01:42,989 --> 00:01:48,751 And so if I understand correctly, we generate a lot more code due to AI systems. 33 00:01:48,751 --> 00:01:53,689 And essentially you were looking for a system that would reduce the burden of the code review mechanism. 34 00:01:53,650 --> 00:01:55,670 Yes, and I think it's twofold. 35 00:01:55,670 --> 00:02:01,933 Indeed, it removes the burden but on the other hand, you will always miss things as a human, 36 00:02:01,933 --> 00:02:08,111 it's easy to overlook something, especially if you don't know the repo, and you only have a few people to get things done. 37 00:02:08,111 --> 00:02:12,124 The chance that you miss a mistake, even a small one, it's very big. 38 00:02:12,124 --> 00:02:17,909 So I think that's another benefit next to the burden itself of doing it. 39 00:02:17,909 --> 00:02:20,277 So also increasing the quality of the reviews. 40 00:02:20,277 --> 00:02:21,644 Yes, yes, 41 00:02:21,644 --> 00:02:27,632 If I look at my own experience, what I personally like in pull request reviews is if they go beyond a code review. 42 00:02:27,632 --> 00:02:29,463 to me this is beyond coding. 43 00:02:29,463 --> 00:02:31,466 This is also about checking the logic. 44 00:02:31,466 --> 00:02:32,258 checking whether... 45 00:02:32,258 --> 00:02:36,982 everything works, doing a local test with the code and giving feedback on that also on the design. 46 00:02:36,982 --> 00:02:42,957 quite often I do see at companies that there's lots of rubber stamping, like people just approving stuff 47 00:02:42,957 --> 00:02:51,082 so I do think there is a huge chance to optimize this workflow to have better reviews So I'm very looking forward to what you're going to show me. 48 00:02:51,082 --> 00:02:53,316 So I would say let's dive in. 49 00:02:53,295 --> 00:02:55,858 I've looked at a few tools that I can test out. 50 00:02:55,858 --> 00:03:02,489 the space is quite big already, so it's really growing, And here are a few of the big players out there. 51 00:03:02,489 --> 00:03:04,704 There is Code Rabbit, which we will take a look at. 52 00:03:04,704 --> 00:03:05,826 There is Qodo, 53 00:03:05,826 --> 00:03:10,491 We have also Greptile, Sourcery, Beto and quite a few others. 54 00:03:10,491 --> 00:03:18,614 we'll try to do just two different ones, but I, in general, validating AI seems always as a hard task, at least to me. 55 00:03:18,614 --> 00:03:19,760 Yeah, indeed. 56 00:03:19,760 --> 00:03:25,148 There's a few benchmarks you could use, but I think it would be good if we could already get a feeling on how these things work. 57 00:03:25,148 --> 00:03:25,980 Yes, indeed. 58 00:03:25,980 --> 00:03:27,962 So let's just dive into the first. 59 00:03:27,962 --> 00:03:34,489 So what I've been doing, Jonny, is this project It's called the Sandi Insights Project. 60 00:03:34,489 --> 00:03:37,672 It's a little internal project for the people listening or watching. 61 00:03:37,672 --> 00:03:41,596 And Jonny asked to review something, to do a few changes 62 00:03:41,596 --> 00:03:52,603 uh Me, not knowing the repo, I started vibe coding, So I asked some feedback of what it was, I checked, okay, what do need to change, where, made a few changes. 63 00:03:52,603 --> 00:03:53,806 I have cursor. 64 00:03:53,806 --> 00:03:57,917 it's VS Code with an extra layer with AI integration basically. 65 00:03:57,917 --> 00:04:00,418 And one of the extensions I installed is CodeRabbit. 66 00:04:00,418 --> 00:04:04,530 So in here you see I'm logged in as myself on GitHub. 67 00:04:04,530 --> 00:04:07,163 and I installed the CodeRabbit extension. 68 00:04:07,163 --> 00:04:12,501 And in here I can see I'm on a separate branch and I can compare to main. 69 00:04:12,501 --> 00:04:14,382 And in here I can review all my changes. 70 00:04:14,382 --> 00:04:17,997 So I will already start that because it takes a while. 71 00:04:17,997 --> 00:04:20,668 It's setting up, reviewing the changes, et cetera. 72 00:04:21,109 --> 00:04:21,474 And... 73 00:04:21,474 --> 00:04:32,119 made a branch and then you, you vibe coded some things or you programmed yourself and now you have the chance to let it review the pull request already that you could create. 74 00:04:32,119 --> 00:04:35,264 This is not a pull request yet, this is just the two branches that we'll compare. 75 00:04:35,264 --> 00:04:36,416 Yes, very good point. 76 00:04:36,416 --> 00:04:38,410 So indeed it's not yet a pull request. 77 00:04:38,410 --> 00:04:40,184 I will make a pull request right now. 78 00:04:40,184 --> 00:04:42,187 I don't think I have any local changes. 79 00:04:42,187 --> 00:04:43,250 Everything is up to date. 80 00:04:43,250 --> 00:04:48,722 So let me quickly go to GitHub while we're waiting for the review anyway. 81 00:04:48,722 --> 00:04:52,421 So this is the code base. 82 00:04:52,421 --> 00:04:54,844 So let's contribute, let's open a request. 83 00:04:54,844 --> 00:04:56,585 I will just create a pull request for now. 84 00:04:56,585 --> 00:05:01,660 I know Jonny in the future, I would like to add some extra descriptions, But for now I'll just create a pull request. 85 00:05:01,660 --> 00:05:06,084 um For once this is okay, I'm allowed. 86 00:05:06,084 --> 00:05:07,977 So let's go back to CodeRabbit. 87 00:05:07,977 --> 00:05:10,649 So CodeRabbit has been very helpful. 88 00:05:10,649 --> 00:05:15,251 The rabbit has been working and you see here quite a few things popping up. 89 00:05:15,251 --> 00:05:19,113 So again, we're looking at a certain branch and we're comparing it to main. 90 00:05:19,113 --> 00:05:23,386 it's checked 12 files and it's detected nine issues. 91 00:05:23,386 --> 00:05:28,068 So what you can do is either I go here through the comments or you click here on the files. 92 00:05:28,068 --> 00:05:28,529 So 93 00:05:28,529 --> 00:05:33,209 This is the biggest file I've changed to the file with the most changes of myself. 94 00:05:33,209 --> 00:05:35,040 So there's a potential issue. 95 00:05:35,040 --> 00:05:37,402 So I can say, click on it. 96 00:05:37,402 --> 00:05:39,653 And that brings me to the actual line question. 97 00:05:39,653 --> 00:05:43,524 So here it tells me there's a type mismatch for origin column prefill. 98 00:05:43,524 --> 00:05:47,804 So the multiselect column says it's a streamlit column type. 99 00:05:47,804 --> 00:05:54,115 So it's a streamlit application and the multiselect column allows you to select multiple aspects. 100 00:05:54,115 --> 00:05:55,486 So the prefill here. 101 00:05:55,486 --> 00:06:03,172 is currently set on an empty string, but that does not make sense since the default value should be a list because it can be multiple values. 102 00:06:03,172 --> 00:06:09,565 that's CodeRabbit is complaining and it seems to me that this is actually a valid remark. 103 00:06:09,565 --> 00:06:16,959 I could do is indeed look this up and validate with actual documentation of Streamlit that this is actually the case. 104 00:06:16,959 --> 00:06:18,331 So let's assume 105 00:06:18,331 --> 00:06:20,603 that the AI is correct in this regard. 106 00:06:20,603 --> 00:06:28,859 Then I can either apply the suggested change, fix with AI, or I can also ignore when I think it's not relevant in this case. 107 00:06:28,859 --> 00:06:30,781 So let's fix it with AI. 108 00:06:30,781 --> 00:06:32,453 And something very cool is happening. 109 00:06:32,453 --> 00:06:34,311 I don't know if you saw what happened, but. 110 00:06:34,311 --> 00:06:36,220 popped up, it opened up a new chat. 111 00:06:36,220 --> 00:06:37,260 Yes, exactly. 112 00:06:37,260 --> 00:06:46,025 So in Cursor, it directly start a new chat with in this file, around lines, blah, blah, blah. 113 00:06:46,025 --> 00:06:47,676 And then it says actually what it found. 114 00:06:47,676 --> 00:06:53,087 So it gives a bit of context where it should find it and then it gives the error. 115 00:06:53,087 --> 00:06:55,788 So this allows me to fix it with AI. 116 00:06:55,788 --> 00:06:58,421 So you basically had these different options. 117 00:06:58,421 --> 00:07:07,952 The first one was actually the diff that it proposed to directly apply it, which is already suggested by CodeRabbit, but you chose the option to fix with AI, which then 118 00:07:07,952 --> 00:07:16,512 transfers this request to fix to a chat window for Cursor, which can then use different models and maybe more powerful things to fix it. 119 00:07:16,512 --> 00:07:17,213 Exactly. 120 00:07:17,213 --> 00:07:24,110 So I'm not sure how this indicates, for example, to co-pilot or if you're using another integration within VS code. 121 00:07:24,110 --> 00:07:28,072 But for example, for Cursor, it works out of the box, So here. 122 00:07:28,072 --> 00:07:35,974 you opt for this AI system to kick in like here with the supplied context versus just pressing the apply now button? 123 00:07:35,974 --> 00:07:37,005 Yes, good question. 124 00:07:37,005 --> 00:07:45,068 So I think applying now, you should always and only do when It is actually correct and you agree with it and you say, yes, this is indeed much better. 125 00:07:45,068 --> 00:07:48,624 A simple rename here, indeed a simple default value. 126 00:07:48,624 --> 00:07:53,981 Either this is my research and I ask, is this the case? 127 00:07:53,981 --> 00:07:56,265 And I basically ask for a second opinion here 128 00:07:56,265 --> 00:08:04,580 So I would use this mainly for either you want to ask more context, more feedback, or if you're unsure about the implementation, sometimes 129 00:08:04,580 --> 00:08:06,902 it does say it can confirm the issue. 130 00:08:06,902 --> 00:08:10,366 It knows about the issue from the context that was given to it. 131 00:08:10,366 --> 00:08:15,933 And it will, in this case, confirm that the fix is indeed what Code Rabbit proposed. 132 00:08:15,933 --> 00:08:22,120 Yes, and it even suggested a second edit because somewhere else I seem to have made the same mistake. 133 00:08:22,120 --> 00:08:24,863 So it even went one step above. 134 00:08:24,863 --> 00:08:26,256 Maybe I'll open a second. 135 00:08:26,256 --> 00:08:29,581 which here is nice, it seems like a critical logic error. 136 00:08:29,581 --> 00:08:30,022 So. 137 00:08:30,022 --> 00:08:41,200 Here it starts complaining again about regarding the same column that returns a list of selected values and the conversion expecting it by comma produces incorrect results. 138 00:08:41,200 --> 00:08:48,256 It gives an example, client work training and I expected to split on comma which wouldn't work because then string, etc. 139 00:08:48,256 --> 00:08:51,400 So here it proposes yet another change. 140 00:08:51,400 --> 00:08:57,808 And it also asks to update the error raising because here I use the same logic. 141 00:08:57,808 --> 00:09:00,240 So let's say, okay, I agree with this. 142 00:09:00,240 --> 00:09:01,031 I've checked this. 143 00:09:01,031 --> 00:09:02,001 I validated it. 144 00:09:02,001 --> 00:09:04,344 I tested it myself in my Python terminal. 145 00:09:04,344 --> 00:09:06,447 I can now apply the suggested change. 146 00:09:06,447 --> 00:09:08,608 Whoop, you see something going over. 147 00:09:08,608 --> 00:09:09,860 Looks very fancy. 148 00:09:09,860 --> 00:09:10,581 And then it's done. 149 00:09:10,581 --> 00:09:14,044 I can also unresolve the comment and go in again. 150 00:09:14,044 --> 00:09:16,385 But now it changed a few lines. 151 00:09:16,385 --> 00:09:21,931 So you can here again, see the changes it made from the test strip, And that's fixed it. 152 00:09:21,931 --> 00:09:23,572 And now can go all through the rest. 153 00:09:23,572 --> 00:09:24,824 I can either ignore. 154 00:09:24,824 --> 00:09:29,023 If I say, well, this issue, I don't think this is an issue, I press ignore. 155 00:09:29,023 --> 00:09:30,370 And then you go through the list. 156 00:09:30,370 --> 00:09:33,889 I do it actually one by one on my own and I go through this. 157 00:09:33,889 --> 00:09:38,213 What CodeRabbit can also do is if you take a look at the website, 158 00:09:38,213 --> 00:09:44,979 next to CLI work, which is actually what I did now, you can also comment on PRs itself. 159 00:09:44,979 --> 00:09:48,282 So what I did now is actually, I call it a pre-flight check. 160 00:09:48,282 --> 00:09:54,744 So basically before I even create the PR, I check myself locally, well, did I do anything wrong or whatever? 161 00:09:54,744 --> 00:09:59,467 But if I make the PR, if you have it enabled and pay for it, 162 00:09:59,467 --> 00:10:02,019 then it can actually do it in the UI itself. 163 00:10:02,019 --> 00:10:07,925 So here you can see a few screenshots and commenting on PR itself, which is really cool. 164 00:10:07,925 --> 00:10:12,659 So you have two places in your whole development process where you can run this app. 165 00:10:12,659 --> 00:10:16,322 So before you're creating a PR and after creating the PR, 166 00:10:16,322 --> 00:10:20,979 And for the PRs, then it needs a GitHub integration, suppose, or Bitbucket or whatever you use. 167 00:10:20,979 --> 00:10:27,064 So you need to allow access to your GitHub repo to make changes. 168 00:10:27,209 --> 00:10:32,042 It's essentially a bot, assume that just visits your PR and then gives comments. 169 00:10:32,621 --> 00:10:40,929 And what I wonder is what you showed locally, Does it only consider code that was committed on that branch or does it also consider uncommitted code? 170 00:10:40,929 --> 00:10:41,750 Good question. 171 00:10:41,750 --> 00:10:44,202 So it basically, there is your answer. 172 00:10:44,202 --> 00:10:48,926 It can review uncommitted changes, committed changes or all changes, which is quite useful indeed. 173 00:10:48,926 --> 00:10:51,999 So you can even before you commit the change. 174 00:10:51,999 --> 00:11:03,298 So even if you do test-driven development and you're very meticulous in the way you never commit non-working code well, this is a nice feature that it even reviews uncommitted 175 00:11:03,298 --> 00:11:04,280 changes before you. 176 00:11:04,280 --> 00:11:11,633 And so the comments that are generated, Are these all applicable only to the changes that you've made or does it go beyond that? 177 00:11:11,633 --> 00:11:14,696 Cause there could be a bug in your main code already. 178 00:11:14,696 --> 00:11:15,918 will simply ignore 179 00:11:15,918 --> 00:11:27,827 Yeah, so it will ignore code that's already there in the sense of it will not review it, but it will look for it as but it will not assume say, hey, last year or last month you 180 00:11:27,827 --> 00:11:32,294 made a change to this or looks like the method is not really correct. 181 00:11:32,294 --> 00:11:33,436 it will not comment on it. 182 00:11:33,436 --> 00:11:35,561 So it's really only on your changes. 183 00:11:35,561 --> 00:11:36,242 Yeah, okay. 184 00:11:36,242 --> 00:11:39,686 I also see there in the pro version, you can use coding guidelines. 185 00:11:39,686 --> 00:11:41,129 That sounds very interesting to me. 186 00:11:41,129 --> 00:11:46,257 Like if you have specific standards, if you can ship them and the review also checks that. 187 00:11:46,257 --> 00:11:48,158 Yeah, I think this is very useful. 188 00:11:48,158 --> 00:11:55,832 also, there is a lot of options as well in AI assets and coding, both with GitHub Copilot, with Cursor, with Joony, whatever you use. 189 00:11:55,832 --> 00:12:04,784 I think all products now have an option to give you guidelines, That in combination with pre-commit hooks that allow you to automatically do other checks as well. 190 00:12:04,784 --> 00:12:05,436 I think... 191 00:12:05,436 --> 00:12:09,256 It also limits the need of this, but it is useful nonetheless. 192 00:12:09,256 --> 00:12:12,868 So this is something I use myself actually. 193 00:12:12,868 --> 00:12:18,599 So I highly recommend testing it out for yourself to see if it works. 194 00:12:18,599 --> 00:12:24,624 For a repo I know pretty well it can be that 50 % or more or 80 % of the comments can disregard. 195 00:12:24,624 --> 00:12:32,415 Basically because I have enough context about what's going on I can say I can ignore this and this and this because I'm quite sure 196 00:12:32,415 --> 00:12:39,200 It's an overkill what they suggest, I think it's not needed or too defensive programming in that case and then I just ignore the comments. 197 00:12:39,200 --> 00:12:40,300 But that's interesting. 198 00:12:40,300 --> 00:12:45,331 that means like 80 % of the comments are not really useful for you personally. 199 00:12:45,331 --> 00:12:57,274 So it depends, in this case, I think it was very useful because I vibe-coded quite a lot the first two comments seemed very plausible and I don't understand the repo well. 200 00:12:57,274 --> 00:13:11,075 If I have a repo where I handwritten, so to say the code base 95 % or 90%, then often I feel the tendency that indeed 80 % of the comments is, 201 00:13:11,075 --> 00:13:21,140 not really that useful, but I have got, for example, few variables that I used the wrong variable name for example, which got through a unit test, And if he didn't catch it, it 202 00:13:21,140 --> 00:13:22,361 would have been a bug later on. 203 00:13:22,361 --> 00:13:27,676 So I actually caught a few bugs which would have caused at least some trouble in production with this. 204 00:13:27,676 --> 00:13:33,565 I do want to dive in maybe on one of these issues, like there's this PII exposure, always interesting. 205 00:13:33,565 --> 00:13:36,060 So I'm actually curious what it suggests there. 206 00:13:36,060 --> 00:13:38,216 Remove or mask the email address from the UI. 207 00:13:38,216 --> 00:13:42,399 Yes, so this is a point which doesn't make that much sense, right? 208 00:13:42,399 --> 00:13:49,240 So I added this as a debug information that I am logged in as hannes.desmet[at]dataminded.com 209 00:13:49,240 --> 00:13:50,620 is that personal information? 210 00:13:50,620 --> 00:13:54,480 Yes, but it's displayed only for me. 211 00:13:54,480 --> 00:14:00,992 So this doesn't make that much sense because it's in the session state of the current user shows you the user email. 212 00:14:00,992 --> 00:14:04,303 So does it make sense that it's actually a PII issue? 213 00:14:04,303 --> 00:14:05,934 I don't think so. 214 00:14:05,934 --> 00:14:08,463 So in this case, I would ignore it. 215 00:14:08,463 --> 00:14:09,010 Yeah, indeed. 216 00:14:09,010 --> 00:14:15,342 And the solution gives us like, it says remove this from the UI, which I can see it does there. 217 00:14:15,342 --> 00:14:21,305 It only shows the number of uh user answers and it also adds something extra in the if test. 218 00:14:21,305 --> 00:14:23,666 It also adds like, and user answers. 219 00:14:23,666 --> 00:14:33,700 So when I saw this, had to actually extend my mental context because I was looking for ah It will just remove the email, but it actually added a few things to the test as well. 220 00:14:33,700 --> 00:14:35,052 which I had to think about. 221 00:14:35,052 --> 00:14:40,806 I would prefer if it gave me a bit more explanation on what it did there, but it seems to be correct. 222 00:14:40,806 --> 00:14:42,801 And then it gives you a second option. 223 00:14:42,801 --> 00:14:49,986 So it says you can mask it and it gives you a way of masking the email address, which I also like that it does that. 224 00:14:50,427 --> 00:14:55,262 But now we don't get an in place button, like apply this, you get different options I see. 225 00:14:55,262 --> 00:15:00,296 Yes, so it's not always there, the button of automatically accepting. 226 00:15:00,296 --> 00:15:05,412 So sometimes you have to do it yourself, either copy pasting it and changing it, or just manually typing it. 227 00:15:05,412 --> 00:15:09,484 sometimes it makes you think also about other aspects, which is quite nice. 228 00:15:09,484 --> 00:15:16,121 If you, maybe you wouldn't just thought of the PII data or you wouldn't think of security logging 229 00:15:16,121 --> 00:15:26,287 tool I tested out as well, which I can't unfortunately anymore, which is quite interesting and I would like to mention as well is um Sourcery So Sourcery is also a code quality 230 00:15:26,287 --> 00:15:26,899 tool. 231 00:15:26,899 --> 00:15:30,941 And it also states it does code review for the AI era. 232 00:15:30,941 --> 00:15:33,713 So it does also security scanning and team. 233 00:15:33,713 --> 00:15:35,945 analytics, not sure what that means. 234 00:15:35,945 --> 00:15:40,029 But the code review, very interesting, is free for open source projects. 235 00:15:40,029 --> 00:15:50,200 So it also comments on the PR itself, like in in GitHub, but it's also a tool you can use in your IDE as code review, So it also works on 236 00:15:50,200 --> 00:15:54,745 both places in your development flow, And I also got quite nice results. 237 00:15:54,745 --> 00:16:04,223 And the nice thing about Sourcery, at least with the comments here in the IDE was that my Cursor could actually see them all. 238 00:16:04,223 --> 00:16:06,705 It had the context of all the comments it posted. 239 00:16:06,705 --> 00:16:11,959 I could just start typing and it could see all the comments, and it integrated really well with Cursor, which was quite nice. 240 00:16:11,959 --> 00:16:17,263 if you want to use it for private repos, at least in the IDE, you have to take a pro license. 241 00:16:17,263 --> 00:16:20,035 it is another nice option, especially for open source projects. 242 00:16:19,755 --> 00:16:24,098 Is this possible to demo it on one of our public repositories? 243 00:16:24,098 --> 00:16:27,479 Yes, I think so. 244 00:16:27,479 --> 00:16:29,439 Live, we can try that. 245 00:16:31,481 --> 00:16:32,109 we have the technology explorations repository, which is public. 246 00:16:32,109 --> 00:16:32,804 Yeah, that one. 247 00:16:32,804 --> 00:16:34,315 technology explorations. 248 00:16:34,315 --> 00:16:35,359 Opening new window. 249 00:16:35,359 --> 00:16:37,543 I don't know if the files are already... 250 00:16:38,727 --> 00:16:39,788 Plugin. 251 00:16:40,749 --> 00:16:42,243 Yes, the extension is installed. 252 00:16:42,243 --> 00:16:44,013 I am currently... 253 00:16:44,013 --> 00:16:45,721 I don't know... 254 00:16:46,224 --> 00:16:47,388 Yeah, I think so. 255 00:16:47,388 --> 00:16:48,261 compare with main 256 00:16:48,261 --> 00:16:49,503 here it gave an overview. 257 00:16:49,503 --> 00:16:59,818 There's one general issue found, the issue division by zero in percentage calculation, which is one of the most default risks you can have, think, in programming. 258 00:16:59,818 --> 00:17:06,275 But at least it is indeed a possible, if your amount of employees is zero, you have an issue, 259 00:17:06,275 --> 00:17:14,887 Here is the comment, so if I hover over it, the blue squiggly line is actually the problem. 260 00:17:14,887 --> 00:17:18,198 So I don't know how it integrates with it. 261 00:17:18,198 --> 00:17:25,181 But here it says on both lines, because it's two lines, it takes it together again I can fix it in chat. 262 00:17:25,181 --> 00:17:25,746 Enter. 263 00:17:25,746 --> 00:17:30,174 in terms of comparison to CodeRabbit, CodeRabbit gave already a proposal quite often. 264 00:17:30,174 --> 00:17:32,377 It says like, this is what you can do to change. 265 00:17:32,377 --> 00:17:36,334 Here we don't get a proposal and we need to use the chat to actually get a fix. 266 00:17:36,334 --> 00:17:38,236 Yeah, so here, cursor 267 00:17:38,236 --> 00:17:39,916 So it made the changes for me. 268 00:17:39,916 --> 00:17:45,610 Again, down here it seems to be the same issue, which wasn't getting flagged. 269 00:17:45,610 --> 00:17:52,652 I don't see anything saying about lines 81 to 83, but cursor seemed smart enough to fix that. 270 00:17:52,652 --> 00:17:53,623 A suggestion. 271 00:17:53,623 --> 00:17:55,051 for some dates. 272 00:17:55,051 --> 00:17:57,857 And here it does give a potential change. 273 00:17:57,857 --> 00:17:59,300 give proposed changes. 274 00:17:59,300 --> 00:18:01,405 But again, I think I have to... 275 00:18:01,405 --> 00:18:02,027 Quick fix. 276 00:18:02,027 --> 00:18:05,816 So here if I do the quick fix, I can actually apply the fix - baf. 277 00:18:05,816 --> 00:18:08,173 This was confusing, there were many quick fixes. 278 00:18:08,173 --> 00:18:09,368 Yes, so it seems to be... 279 00:18:09,368 --> 00:18:13,233 I'm not sure why, it is a bit confusing also targeting the same... 280 00:18:14,276 --> 00:18:16,878 No, I think it's always duplicate it seems. 281 00:18:16,878 --> 00:18:18,685 Yes, see it's always... 282 00:18:18,685 --> 00:18:24,432 And if we look at this compared to Code Rabbit, it seems like Code Rabbit was actually a bit more advanced. 283 00:18:24,432 --> 00:18:29,319 In general, I thought the integration was better. 284 00:18:29,319 --> 00:18:33,482 the sense that I don't know if CodeRabbit by then had to fix it with AI button. 285 00:18:33,482 --> 00:18:34,855 I don't think it had it. 286 00:18:34,855 --> 00:18:41,892 So it was back then only the quick fix or ignore but not a fix with AI. 287 00:18:42,413 --> 00:18:46,497 Which that was the part I really liked about the integration. 288 00:18:46,497 --> 00:18:48,390 One cool thing it had, 289 00:18:48,390 --> 00:18:51,634 It had the option to create diagrams, which was very cool. 290 00:18:51,634 --> 00:18:53,895 So it has the context of the open file. 291 00:18:55,578 --> 00:18:57,791 the UI is maybe not the best. 292 00:18:57,791 --> 00:18:58,813 Okay, nice. 293 00:18:58,813 --> 00:19:00,127 So that is quite cool. 294 00:19:00,127 --> 00:19:02,849 this one is actually quite useful, I think. 295 00:19:03,090 --> 00:19:07,115 I mean, it's still very minimal, but I can imagine what it could do. 296 00:19:07,115 --> 00:19:08,828 Yeah, I really liked the diagram. 297 00:19:08,828 --> 00:19:14,152 this was quite fun to play with, but I don't know how useful is. 298 00:19:14,070 --> 00:19:17,280 Do you have a rough idea on the cost and the cost models? 299 00:19:17,280 --> 00:19:20,073 CodeRabbit it's per developer as well. 300 00:19:20,073 --> 00:19:29,283 Sourcery pricing, let's quickly check, Pro, seat So all it's per seat so depending on the size of your organization, except if it's enterprise, 301 00:19:29,283 --> 00:19:36,503 Okay, so they're all charging per seat and it's roughly in the 20, 30 dollar range, more or less. 302 00:19:36,494 --> 00:19:41,311 But so if I understand correctly, CodeRabbit and Sourcery are quite similar. 303 00:19:41,311 --> 00:19:43,363 They have this cursor integration. 304 00:19:43,363 --> 00:19:46,227 They also have this dashboard and GitHub integration. 305 00:19:46,227 --> 00:19:48,220 So they're direct competitors. 306 00:19:48,220 --> 00:19:49,741 Yeah, yeah, exactly. 307 00:19:49,741 --> 00:19:52,343 I think CodeRabbit is the biggest player out there in general. 308 00:19:52,343 --> 00:19:59,226 I think this, to be fair, having this for free in open source projects, yeah, count me in I would say. 309 00:19:59,295 --> 00:20:05,261 I am curious to see at least in the space we're in, if it's gonna change. 310 00:20:05,787 --> 00:20:05,399 Yeah. 311 00:20:05,554 --> 00:20:07,657 and how much it will improve. 312 00:20:07,657 --> 00:20:15,996 because if you need context over different repos, then it becomes, I think, very difficult for AI, at between, especially in data engineering, platform engineering. 313 00:20:15,996 --> 00:20:19,420 There's a lot of integrations, a lot of glue between components. 314 00:20:19,420 --> 00:20:21,643 You still need to have that context. 315 00:20:21,643 --> 00:20:24,764 You mean also between different projects and repositories? 316 00:20:25,452 --> 00:20:26,732 but we have a video for that. 317 00:20:26,732 --> 00:20:34,016 Emil demoed Cursor workspaces, which allows you to go beyond your project and beyond your code base. 318 00:20:34,083 --> 00:20:37,830 Yeah, that's very well hidden promotion. 319 00:20:37,830 --> 00:20:39,472 It's good, it's good. 320 00:20:39,472 --> 00:20:41,266 I like it, I will check it out actually. 321 00:20:41,250 --> 00:20:47,536 How would you summarize your experience with all these AI code assistants or pull request assistants? 322 00:20:47,536 --> 00:20:50,209 I would say, first of all, it was fun playing around with. 323 00:20:50,209 --> 00:20:55,037 So if you are curious, try it out for yourself, see what works. 324 00:20:55,037 --> 00:20:59,593 If you have a team that maybe wants to use it, at least give it a chance. 325 00:20:59,593 --> 00:21:00,825 find out what works for you. 326 00:21:00,825 --> 00:21:02,108 Is it a preflight check? 327 00:21:02,108 --> 00:21:08,221 Is it having both Sourcery and CodeRabbit on the same repo and doing it all? 328 00:21:08,221 --> 00:21:09,895 go ahead and play with it. 329 00:21:09,895 --> 00:21:10,326 Yeah. 330 00:21:10,326 --> 00:21:11,246 Okay. 331 00:21:11,246 --> 00:21:16,191 My takeaway on all of this is basically you showed me that you have two variants. 332 00:21:16,191 --> 00:21:20,986 You have this local variant and the preflight check is always a good idea before you push. 333 00:21:20,986 --> 00:21:28,616 And then just like with pre-commit hooks, you could also run something on your PRs so that everybody can benefit from that in a centralized way. 334 00:21:28,616 --> 00:21:30,087 We saw several different tools. 335 00:21:30,087 --> 00:21:33,381 The space there is getting a bit crowded it seems. 336 00:21:33,381 --> 00:21:38,728 I think this is very nice to automate also some review systems and to get feedback fast. 337 00:21:39,248 --> 00:21:39,901 Yes. 338 00:21:39,901 --> 00:21:44,686 And one thing I might forgot to say is that I showed you two UI integrations. 339 00:21:44,686 --> 00:21:47,907 So both Sourcery and CodeRabbit here via UI extension. 340 00:21:47,907 --> 00:21:51,580 But for example, CodeRabbit released, a CLI tool. 341 00:21:51,580 --> 00:21:55,543 So you can also use it in the CLI, which makes, for example, I'm also a PyCharm user. 342 00:21:55,543 --> 00:21:57,033 I don't only use Cursor. 343 00:21:57,033 --> 00:22:01,246 And there you don't have an extension for CodeRabbit, so that it becomes a bit more difficult. 344 00:22:01,246 --> 00:22:04,940 So there I could, for example, use the CLI, which is maybe a nice thing to know. 345 00:22:04,940 --> 00:22:05,943 All right, good. 346 00:22:05,943 --> 00:22:11,523 Hannes, thanks a lot for showing us around in the space of the AI pull request reviewers. 347 00:22:11,523 --> 00:22:12,771 Very happy to do so. 348 00:22:12,771 --> 00:22:14,264 All right, let's wrap it up. 349 00:22:14,264 --> 00:22:15,998 I want to thank everybody for watching. 350 00:22:15,998 --> 00:22:18,032 Thank you all and I hope to see you next time. 351 00:22:18,032 --> 00:22:18,737 Bye bye. 352 00:22:18,737 --> 00:22:19,585 See ya.