Episode 519: The Last Frontier: Can AI Finally Conquer BC Report Layouts?

WEBVTT
00:00:00.719 --> 00:00:06.959
So I think I mean we already know a few guys, right, that are now telling everybody how to do stuff.
00:00:07.280 --> 00:00:12.800
But I think each organization has to get have someone like that that will pioneer.
00:00:15.519 --> 00:00:19.359
Welcome everyone to another episode of Dynamics Corner.
00:00:19.679 --> 00:00:23.120
When you get started with AI, do you use all the tokens?
00:00:23.359 --> 00:00:25.760
Or are you a little bit more stingy?
00:00:26.000 --> 00:00:27.039
Because I am sometimes.
00:00:27.199 --> 00:00:28.640
I'm your co-host, Chris.
00:00:28.879 --> 00:00:29.679
And this is Brad.
00:00:29.760 --> 00:00:32.799
This episode is recorded on May 6th, 2026.
00:00:32.960 --> 00:00:34.399
Chris, Chris, Chris.
00:00:34.640 --> 00:00:46.399
Here we are, fresh back from Directions North America with another episode talking about how you can use AI within your business central environment and your development workflow and process and other things.
00:00:46.479 --> 00:00:51.119
With us today, we had the opportunity to speak with another great Microsoft MVP, Bill John.
00:01:08.560 --> 00:01:09.200
That is right.
00:01:09.359 --> 00:01:10.319
Lack asleep.
00:01:10.560 --> 00:01:12.079
Yeah, well, that's not good for you.
00:01:12.159 --> 00:01:13.920
It's not good for you to have lack of sleep.
00:01:14.000 --> 00:01:14.480
Good afternoon.
00:01:14.560 --> 00:01:15.439
How are you doing?
00:01:16.159 --> 00:01:17.120
Hi, doing well.
00:01:17.200 --> 00:01:18.000
You too, guys.
00:01:18.319 --> 00:01:19.439
Excellent, excellent.
00:01:20.000 --> 00:01:26.640
It was nice to see you last week uh over at the Directions North America conference.
00:01:27.439 --> 00:01:30.719
Yeah, it was quite an interesting conference, I would say.
00:01:31.200 --> 00:01:34.319
Quite uh new stuff coming around.
00:01:35.359 --> 00:01:35.840
Yes.
00:01:36.000 --> 00:01:36.239
Yes.
00:01:36.400 --> 00:01:39.200
Was that your first time going to that conference?
00:01:39.680 --> 00:01:42.959
Yeah, it was actually my first time to the US in general.
00:01:43.200 --> 00:01:45.200
So everything new for me.
00:01:47.519 --> 00:01:51.359
I hope you were able to get out while you were here and explore.
00:01:51.760 --> 00:01:53.920
Well, at least it's at least it's Orlando, right?
00:01:54.000 --> 00:01:58.799
The first time, at least they've got the good the good weather, places to go.
00:01:59.280 --> 00:02:01.280
So did you enjoy that?
00:02:03.040 --> 00:02:03.280
Yeah, yeah.
00:02:03.439 --> 00:02:10.319
I mean it it's um it's a bit, you know, whenever you go to a country first, you you see the the different mentality.
00:02:10.479 --> 00:02:15.439
That's always kind of refreshing for me to see how the people think on the other part of the world.
00:02:15.520 --> 00:02:20.960
But yeah, I also went to Kennedy Center to I also managed to see the rocket launch.
00:02:21.439 --> 00:02:22.240
Oh, that's right.
00:02:22.479 --> 00:02:27.039
I remember I remember when we were walking into the event uh hearing about that.
00:02:27.199 --> 00:02:28.319
That's exciting.
00:02:28.639 --> 00:02:34.000
You you hit on something that was inter that was a key point.
00:02:34.159 --> 00:02:43.520
Uh one time I noticed I went over to Iceland and I was meeting with a group of individuals from many different countries.
00:02:43.680 --> 00:02:45.759
Uh, we went over there for a meeting.
00:02:46.080 --> 00:03:08.080
And as you had mentioned, the it's the mentality or the difference, it's it's great to get the perspective of seeing a different culture or talking with people of different cultures in person versus sometimes what you may see or hear because of what you may see online, what you may see on the news, or what you may see somewhere else.
00:03:08.400 --> 00:03:14.319
So um it it's it is, it's refreshing to see what that is like.
00:03:14.639 --> 00:03:16.960
Uh without saying too much, did it match?
00:03:17.039 --> 00:03:21.919
Did it all match your expectations of what you thought, or was it anything different?
00:03:22.080 --> 00:03:23.520
And you don't have to say if it's good or bad.
00:03:23.599 --> 00:03:24.960
I'm not asking for that.
00:03:25.120 --> 00:03:27.520
Just did it match your expectations?
00:03:28.080 --> 00:03:33.120
Yeah, I didn't have that many expectations, except for what I have seen from movies.
00:03:33.759 --> 00:03:38.319
But I I would say it's matched for the most part, I would say.
00:03:38.400 --> 00:03:38.639
Yeah.
00:03:38.879 --> 00:03:39.439
So it's good.
00:03:39.759 --> 00:03:42.400
It's exciting that you were able to go see the rocket.
00:03:42.800 --> 00:03:47.360
I if I had known, I would have taken a quick ride over there as well, too, to see the rocket.
00:03:47.439 --> 00:03:51.039
Uh I know for you guys, I remember talking to you that night, it was you didn't know.
00:03:51.120 --> 00:03:54.400
You were just going over to sea, and you happened to see the rocket launch.
00:03:54.479 --> 00:03:55.520
That must have been impressive.
00:03:55.840 --> 00:04:02.080
Yeah, we just went and like five minutes in the uh after there was like a rocket launch.
00:04:02.159 --> 00:04:04.240
So yeah, didn't expect that.
00:04:04.560 --> 00:04:08.319
No, it's uh it's something and you can feel it as it goes all through you, right?
00:04:08.479 --> 00:04:10.400
It's were you close enough to feel it?
00:04:10.479 --> 00:04:24.480
Uh yeah, I mean the sound when it comes back, especially then you really uh hear it and it's it's you have no, it's shaking a bit, you know, because it's uh and and it's it has a delay also, uh also on the monitor, but also with the sound.
00:04:24.639 --> 00:04:28.240
So quite quite interesting, I would say, uh experience.
00:04:29.040 --> 00:04:30.079
That's excellent.
00:04:30.319 --> 00:04:36.480
Uh okay, we want to talk with you a lot about things, but before we get into that conversation, would you mind telling us a little bit about yourself?
00:04:37.519 --> 00:04:43.519
Yeah, um I'm uh business central developer, uh techie guy, so to say.
00:04:43.839 --> 00:04:48.160
Um lately been playing a lot with uh AI stuff.
00:04:48.720 --> 00:04:52.319
Uh was also busy with uh layouts.
00:04:52.560 --> 00:04:57.839
But generally speaking, uh I'm with Business Central since 2008, so quite a while.
00:04:58.079 --> 00:05:08.959
I'm also uh titled as M Mv, meaning I chat quite a bit uh about Business Central.
00:05:09.279 --> 00:05:12.319
And yeah, that's that sums it up, I would say.
00:05:13.120 --> 00:05:13.439
Great.
00:05:13.920 --> 00:05:14.160
Great.
00:05:14.720 --> 00:05:37.839
So I definitely would like to speak with you about the AI and in the development and how it's helping you and some of the things that you're doing, because I think that's it's it's an important part of Business Central now, not only uh with Microsoft adding a lot of AI features to Business Central the last few years and individuals working with AI to help with the implementations, but there's also a lot of individuals working with the development.
00:05:38.079 --> 00:05:42.639
But you had mentioned there's a lot of new things coming out just to take away from the conference.
00:05:42.720 --> 00:05:50.319
Um, what are some of the features or some of the new things within Business Central that are with 2020?
00:05:50.720 --> 00:05:51.199
Where are we?
00:05:51.839 --> 00:05:52.879
I forgot where we were right now.
00:05:53.040 --> 00:05:55.519
2026 wave one perfect for a moment.
00:05:55.600 --> 00:05:56.959
I really thought it was October.
00:05:57.120 --> 00:06:00.560
Like, I don't know what's going on, but I was about to say something like 2025.
00:06:00.959 --> 00:06:07.920
You also have a lack of sleep, but I the you know you know last night I had like the best sleep.
00:06:08.000 --> 00:06:11.120
I had a 91 sleep score, so I'm not I'm refreshed today.
00:06:11.360 --> 00:06:22.720
So um for 2026 wave one, what are some of the exciting features either from the application that you we saw that they had presented or from the development from the technical area?
00:06:22.879 --> 00:06:26.000
Uh I always like to hear individual perspectives.
00:06:26.399 --> 00:06:41.439
Yeah, I think I think I mean uh from technical, there are many stuff from technical, but uh from technical side, but I think generally speaking, what's what's the biggest highlight for me for business central in general is the expense uh management uh agent.
00:06:42.079 --> 00:06:45.920
And I'm looking forward to seeing it and to seeing how it actually works.
00:06:47.120 --> 00:06:49.519
How will it be access accepted?
00:06:49.920 --> 00:06:51.680
Uh but it's an AI feature.
00:06:51.839 --> 00:07:01.759
Uh and I but I think this is the feature that might you know change how people uh use Business Central.
00:07:02.160 --> 00:07:08.079
Uh I know it was developed completely using AI and it was developed at a faster pace than the other features.
00:07:08.560 --> 00:07:16.959
So I'm really wondering how it will be accepted and how it will it work out at the end, you know, in the wild and it's released in the wild when people start using it.
00:07:17.199 --> 00:07:19.519
So that's the top one, I would say.
00:07:19.759 --> 00:07:20.079
Yeah.
00:07:20.480 --> 00:07:23.680
I think what I think to me was expense management.
00:07:23.839 --> 00:07:41.360
I I appreciate you bringing that up, Malik, because to me that was kind of like the almost like the the story that I've been sort of waiting to hear that a solution was released purely created, almost purely created by AI itself.
00:07:41.519 --> 00:07:54.000
Obviously, there's still a human in the loop, but that that to me that was like, okay, I see the capabilities now from an ERP perspective that they release in a very short period of time.
00:07:54.319 --> 00:07:58.879
And it was pretty well done uh in a very short period of time.
00:07:59.040 --> 00:08:02.800
So that to me was like, okay, now we're grooving.
00:08:02.879 --> 00:08:08.399
I mean, and now I see it in live of what they could do with it.
00:08:08.639 --> 00:08:12.399
So that was mind-blowing for me at least when I saw it.
00:08:13.040 --> 00:08:13.759
Yeah, yeah.
00:08:14.000 --> 00:08:28.319
Did they also adopted this new way of I mean, I'm not sure if you're you're contributing actively on Microsoft repos, but they already uh started uh rolling out this workload that they have, like adjanting triad uh code review and everything.
00:08:29.040 --> 00:08:31.439
PC quality, I believe that was the repo.
00:08:31.839 --> 00:08:32.960
PC quality, yeah.
00:08:33.039 --> 00:08:40.080
But even before they released it officially, they started using it on their repos, on on the issues created for their repos.
00:08:40.399 --> 00:08:45.440
And I'm also um kind of keenly waiting to see, you know.
00:08:45.600 --> 00:08:54.240
I mean I already see some interesting stuff like the stuff that gets caught by agents in development process.
00:08:54.480 --> 00:09:01.120
So I'm also interested, you know, interested to see how that will roll out at the partners.
00:09:02.799 --> 00:09:06.240
I I think it may take some time for for the adoption.
00:09:06.320 --> 00:09:18.320
I mean, for some, uh it's it it was the I've had had many conversations with many partners about the technical portions of business central, primarily development, uh, while I was at the event.
00:09:18.559 --> 00:09:25.200
And partners are at different stages of using AI in the development.
00:09:25.679 --> 00:09:30.960
Seeing how the BC quality workflow works, I appreciate that and I like it.
00:09:31.200 --> 00:09:38.240
I have been doing that or had been working with that over the past couple months, a similar process, and it does work well.
00:09:38.879 --> 00:09:46.799
But the key is you have to know what it's doing from a high level at least, and then be able to review it.
00:09:47.039 --> 00:10:06.080
So I like how the BC quality will have the Microsoft, I uh use the word rules or instructions, the community instructions, and then partners can put their own instructions in there as well to help monitor the code that gets in there because you have to make sure that you properly test.
00:10:06.480 --> 00:10:23.679
Just to say that something was developed by AI doesn't mean it was perfect, nor does it mean it's not good, because uh again, Microsoft was um open enough during the event to state that the expense agent was written, I think was 90% by AI, was what they had mentioned, or some number.
00:10:23.759 --> 00:10:32.639
It's still it was still a significant number, and it only the period of time it took to develop was six months in that period of time, or from like idea to release, I think.
00:10:32.720 --> 00:10:36.559
Um that's what I recall from the discussion.
00:10:36.639 --> 00:10:39.360
If I'm incorrect, then it's just my old age.
00:10:39.759 --> 00:10:40.480
I think you're right.
00:10:40.720 --> 00:10:48.159
It was like the the moment they talked about it all on all up to the delivery in terms of the development was like very short as well.
00:10:48.320 --> 00:10:53.840
So I think that's pretty impressive for what they were able to show in a short period of time.
00:10:54.240 --> 00:11:05.120
Yes, and it it does show how you can expedite the development process of features within the ERP or for PTEs for customers.
00:11:05.279 --> 00:11:18.080
Uh, not everything can be done fully by AI at this point, but it's nice with the assisted development and to be able to basically build your own rules with it uh to help with the quality of code, which is something that I think we need.
00:11:18.240 --> 00:11:19.600
Uh yeah, for sure.
00:11:19.759 --> 00:11:20.879
For sure, I would say.
00:11:21.120 --> 00:11:24.799
I mean, looking at the partner code, that's something that's that's desperately needed.
00:11:25.039 --> 00:11:30.000
It's just the question that you already asked is about the adoption base.
00:11:31.360 --> 00:12:06.720
Yeah, yes, it's it's tough, I think, from the partner point of view, because it's almost the equivalent of changing the wheels on the bus while it's moving in some cases, because the partners are working with customers and they're trying to continue through the process that you do have to allocate some uh some time or some resources or some individuals to focus on that to set that up because it's not something you can just say, I'm going to clone the repo, throw this into my workflow, and everything will just be fine.
00:12:06.799 --> 00:12:09.440
Uh, you re uh it's something that really needs to be understood.
00:12:10.000 --> 00:12:11.600
I think that's just that's key, Brent.
00:12:12.320 --> 00:12:13.279
I think that's key.
00:12:13.919 --> 00:12:20.480
I think that's key because like if you uh you know, with with all these features coming out for smaller partners is a little tough.
00:12:20.720 --> 00:12:28.159
So um for uh maybe slightly larger partner, you do have to um invest the time.
00:12:28.320 --> 00:12:34.159
So meaning as a partner, you have to at least have one person and say, okay, this is what you need to learn.
00:12:34.320 --> 00:12:37.360
Give 30 days, maybe, you know, just like building a habit, right?
00:12:37.440 --> 00:12:39.519
So 30 days is all you need to do.
00:12:39.759 --> 00:12:49.440
Help us build a framework, spend that, you know, invest that time, and then come back together and say, okay, where can we at, you know, where can we transition to this workflow?
00:12:49.600 --> 00:12:58.080
Uh I think that's gonna be the struggle because as a partner, you want to, you know, you gotta build your time somewhere, get gonna make some revenue.
00:12:58.320 --> 00:13:00.000
So that's gonna be an interesting balance.
00:13:00.399 --> 00:13:05.519
That's a whole different discussion, but it's it is having individuals focus on it is helpful.
00:13:05.679 --> 00:13:11.840
And uh from what I found, it it has to be collaborative as well because every everybody will have different ideas.
00:13:12.000 --> 00:13:24.080
Uh it's not even ideas like you have to take everyone's ideas, but it's all so new and there's also much that's changing that having different perspectives in the process is important.
00:13:24.320 --> 00:13:31.519
But one thing that I've I I have witnessed firsthand is the consistency in an organization and how you're doing it.
00:13:31.679 --> 00:13:51.519
Because if people are using AI and development in your organization, you want them to follow the same flow, not have someone going down one road and another person going down another road, because then it it becomes I think it becomes more difficult to manage than it would be if everyone's just doing code on their own anyway.
00:13:51.679 --> 00:13:58.559
I think it I think it magnifies and accelerates inconsistent processes if you don't have a standard workflow.
00:13:58.799 --> 00:14:00.159
So exactly.
00:14:00.399 --> 00:14:10.720
And I think I think the challenge will be in the following period actually finding out which people are going to be those managing in you know, introducing the processes, right?
00:14:11.519 --> 00:14:17.679
So I think I mean the we already know a few guys, right, that are now telling everybody how to do stuff.
00:14:18.159 --> 00:14:26.240
But I think each organization has to get have someone like that that will pioneer the the way through the adopting adoption.
00:14:26.960 --> 00:14:27.600
Yes.
00:14:28.240 --> 00:14:29.840
And it's changing fast.
00:14:30.080 --> 00:14:32.480
And I wonder uh you you work with it too.
00:14:32.559 --> 00:14:34.080
What are some of the things that you're working with it?
00:14:34.159 --> 00:14:36.240
Because then I have some questions on that as well.
00:14:37.360 --> 00:14:43.360
Uh well primarily I'm I'm I I I'm using the GitHub compiler so far.
00:14:44.399 --> 00:14:48.080
Um I also have played a bit with the plot code.
00:14:48.639 --> 00:15:03.519
And uh I I use it, I mean, currently I'm developing either um stuff that are uh for uh partner, one partner I'm working for, or for another project I'm working on, or for Microsoft actually, as a contribution.
00:15:04.080 --> 00:15:13.519
And in all of those uh different projects, um it's a challenge for me to have the one rule that applies for all of those.
00:15:13.679 --> 00:15:22.399
That's why I like the idea that Christoph uh uh talked about on in Orlando about having these uh how did he call it?
00:15:22.480 --> 00:15:27.440
I forgot, but basically the game that that's gamification.
00:15:28.399 --> 00:15:29.600
Gamified it.
00:15:30.720 --> 00:15:32.159
That blew my mind.
00:15:32.399 --> 00:15:38.799
I I sat with both he and and Tinay, uh Starch as well, and my mind just went.
00:15:40.639 --> 00:15:42.720
Have you started already playing with it?
00:15:42.879 --> 00:15:43.679
Or you played.
00:15:44.080 --> 00:15:45.360
I have it planned out.
00:15:45.440 --> 00:15:49.840
Uh we we just got back uh some days ago and I had to get caught up.
00:15:50.000 --> 00:15:50.960
I think when did we get back?
00:15:51.120 --> 00:15:52.159
Last end of last week?
00:15:52.320 --> 00:15:52.879
This weekend?
00:15:53.440 --> 00:15:53.919
Thursday.
00:15:54.080 --> 00:15:55.919
Yeah, last week, Thursday.
00:15:57.600 --> 00:15:59.360
I'm in a time warp for some reason.
00:15:59.519 --> 00:16:02.879
Uh I do have planned out what I want to do with it.
00:16:03.120 --> 00:16:08.639
And I spoke with both of them, I had the opportunity to spend a lot of time with both of them at the conference as well.
00:16:09.200 --> 00:16:13.120
And they were able to share with me how they architected that.
00:16:13.200 --> 00:16:19.840
And I've plotted out the game that I want to replicate for my youth, and I'm going to try to set that up.
00:16:20.080 --> 00:16:20.320
Yeah.
00:16:20.559 --> 00:16:21.679
So have how about you?
00:16:21.759 --> 00:16:22.799
Have you done anything with that?
00:16:23.039 --> 00:16:23.519
No, no.
00:16:23.600 --> 00:16:26.159
Um, I I I generally have a like a constant feeling.
00:16:26.240 --> 00:16:32.559
Uh my aha moment with AI can came somewhere uh around the last uh end of the last year, actually.
00:16:32.799 --> 00:16:49.440
And I think I'm still uh I mean, I I I think there is still a lot to learn for from my from for myself and and to explore, and I feel like when I hear Christoph or somebody uh like that, I feel I haven't uh I haven't explored enough yet, to be honest.
00:16:49.600 --> 00:17:05.440
Uh and I I also have I'm also attending the Jeremy's uh course on on agent development just to get uh different perspectives and to try different stuff out because currently it feels like I'm just burning all the tokens that I have available.
00:17:05.680 --> 00:17:08.319
I am achieving pretty good results, I would say.
00:17:08.400 --> 00:17:13.759
Uh or let's put it this way my code is less buggy when the AI does it.
00:17:14.000 --> 00:17:21.519
And I think I'm at the mark that probably 95% of my code is written by uh GitHub Copilot, actually.
00:17:21.759 --> 00:17:22.400
That's GitHub.
00:17:22.799 --> 00:17:23.680
That's wild.
00:17:24.000 --> 00:17:25.279
That's crazy.
00:17:26.000 --> 00:17:33.200
With that, you're saying 95% is uh created by GitHub Copilot.
00:17:33.839 --> 00:17:40.400
How did you set are you using co are you using GitHub Copilot straight?
00:17:40.640 --> 00:17:43.680
Or did you use any of instruction files?
00:17:43.839 --> 00:17:46.319
Did you create any instruction files, create any agents, creating?
00:17:46.799 --> 00:17:48.880
Are you using any of the ones that are available?
00:17:49.039 --> 00:17:50.160
A lot of people are doing it.
00:17:50.240 --> 00:17:58.480
And just to go back, I I feel like with you, I have been pretty deep into it since the end of last year.
00:17:59.039 --> 00:18:03.839
And I've been doing a lot of things and working with a lot of things with AI.
00:18:04.000 --> 00:18:11.519
And then I went to the conference as well, and I was pretty, I don't say comfortable, I don't think anyone can be an expert because it changes every day.
00:18:11.680 --> 00:18:13.599
Something new comes out every day in this technology.
00:18:13.759 --> 00:18:16.319
So but I was comfortable with saying, Oh, I have a good understanding.
00:18:16.480 --> 00:18:23.200
And then I was talking with others, like you said, like with KB and with Tina and a few others, and I just went, I don't know anything.
00:18:23.599 --> 00:18:29.440
I really felt like I really felt like I was at the like, I don't even know what I'm doing.
00:18:29.519 --> 00:18:38.960
And I'm I I thought I was doing a lot, but uh the the the point with that is everybody's at a different level and applying it differently, and it is a new technology.
00:18:39.119 --> 00:18:44.240
I will say a year ago, if you hadn't started with it, you you had some time.
00:18:44.319 --> 00:18:52.720
I think we're at the point now this year, if you haven't started with it, you're probably starting to fall behind uh as well.
00:18:53.039 --> 00:18:57.680
But to to go back to the question of how did you set that up um as well.
00:18:58.319 --> 00:19:01.759
Well, I actually uh I haven't used many instruction files.
00:19:01.839 --> 00:19:03.839
I I knew I knew always that it existed.
00:19:04.000 --> 00:19:18.960
So right now, uh I mean, uh as an MVP, I have the access to the Pro Plus version like you do, which means a lot of uh uh tokens, I mean how they call it professional credits or something like that, are available to me.
00:19:19.279 --> 00:19:26.799
So right now I'm burning it completely, uh switching between those um reasoning models and and the normal ones.
00:19:27.039 --> 00:19:31.920
And I don't have much of a setup as I said, I have three different projects, and it's always different challenges.
00:19:32.160 --> 00:19:39.680
So I'm just starting to to explore how to exploit uh sub agents and instructions more.
00:19:40.160 --> 00:19:48.559
But so far I would say I was I was just I wouldn't I wouldn't say excelling, but managing to write uh the my requirements properly.
00:19:48.880 --> 00:19:59.839
And right now I'm in a phase where I figure I'm figuring out okay, sometimes when I when I when I'm developing, I often repeat stuff you know that got wrong during development.
00:20:00.160 --> 00:20:06.000
Maybe I should have now um a sub agent or or a skill for that purpose.
00:20:06.400 --> 00:20:09.759
But I'm not uh I'm not I'm not there yet.
00:20:10.559 --> 00:20:11.839
That's my uh thing.
00:20:11.920 --> 00:20:21.200
So I'm I'm currently using it without much of a setup, and that's what I would recommend everybody who's charting to try without setup because it works actually also without a setup.
00:20:21.759 --> 00:20:31.119
Yeah, it it it it will burn the the the tokens, it will you know you'll sit after the first weekend seeing that you don't have any remaining, but you should shout.
00:20:32.960 --> 00:20:38.000
Yes, you you did say something uh that caught my attention.
00:20:38.480 --> 00:20:46.400
When you start out, start out plain to start working with it, to use it to see what it does, to see what the different models do.
00:20:46.559 --> 00:20:52.559
The instruction files, the skills, the subagents, the agents, uh again, whatever term you want to use, right?
00:20:52.720 --> 00:20:55.119
Because every day I think I hear a new term.
00:20:56.640 --> 00:21:05.599
That's where I personally struggled at first because I saw a lot of things that everybody was doing, and I just felt overwhelmed because I'm like, oh, I have to use this file.
00:21:05.839 --> 00:21:07.680
And I really didn't understand it.
00:21:08.319 --> 00:21:20.240
But by starting base, as you had mentioned, uh you get the point to understand the platform, you get to understand how it works, and then those files seem to start making sense.
00:21:21.359 --> 00:21:29.279
And 95% of your code being done without any of those additional setups, that's rather impressive.
00:21:29.839 --> 00:21:31.839
And it can be done.
00:21:33.359 --> 00:21:34.559
What's what's your statistic?
00:21:34.640 --> 00:21:36.319
How how far are you with with that?
00:21:36.559 --> 00:21:46.240
Have you managed to I am it it I you I'll say I'm almost near a hundred percent.
00:21:46.799 --> 00:21:59.759
And I use the near a hundred percent because sometimes things will come back that don't work properly or look properly, and I make the decision.
00:22:00.240 --> 00:22:01.359
I always review it.
00:22:01.599 --> 00:22:05.839
And I have my workflow is I'll have the tests for it.
00:22:05.920 --> 00:22:09.759
You know, I'll create the I'll I'll plan out what the modification needs to be.
00:22:09.920 --> 00:22:11.759
So I'll go through the whole planning phase.
00:22:11.920 --> 00:22:15.039
Then I'll break it up into smaller issues to complete it.
00:22:15.200 --> 00:22:20.640
And then I'll set the um agents to go through and do the work, right?
00:22:20.720 --> 00:22:21.680
Uh planning.
00:22:21.920 --> 00:22:26.240
I do have a planning agent, I do have a working agent, I do have a reviewing agent.
00:22:26.400 --> 00:22:33.680
Uh I don't I do a lot of the documentation with it, but I don't use an agent for that because within that framework, I still can do it.
00:22:34.240 --> 00:22:51.200
Once the code is there, if I look at it and there's something in what I what I feel is incorrect, and sometimes it's incorrect because it won't even compile, rather than prompt it to fix it, if I could it's faster for me to change it, I'll change it.
00:22:51.359 --> 00:23:06.640
Then depending upon what it is, I may add it to a rule so that hopefully next time, which seems to work, so now it will have the the agent that's working on the task will have a little more experience on what to do.
00:23:06.880 --> 00:23:09.920
Um I got a question for both of you, really quick.
00:23:10.079 --> 00:23:20.880
So when you're mentioning all the agents, Brad, and well if you're using agents as well, for all the little tasks, you know, to say maybe have uh a different models for those agents.
00:23:21.440 --> 00:23:23.599
The last component's always documentation.
00:23:23.759 --> 00:23:28.720
Obviously, you can have an agent create uh documentation, but do you right now, Brad?
00:23:28.799 --> 00:23:33.599
You should you said you may have done a manual documentation in terms of like screenshots.
00:23:33.839 --> 00:23:38.240
Do you know anyone that has used an agent to create screenshots for those?
00:23:38.400 --> 00:23:40.240
Like click here, do that.
00:23:40.400 --> 00:23:42.240
Um, or is that still done manual?
00:23:42.480 --> 00:23:48.079
Because uh obviously you're gonna have to hand it to somebody that's like follow the screenshot kind of deal.
00:23:48.319 --> 00:23:51.680
But I don't know if anyone has used an agent to create the screenshot.
00:23:51.759 --> 00:23:55.359
I'd be surprised, I'd be surprised if anyone, if anyone's goes into this, yeah.
00:23:55.440 --> 00:24:04.000
I have heard somebody, I think when I was in Nordics, Doc uh there's Norish Nordics that was uh two or three uh a month ago.
00:24:04.319 --> 00:24:08.799
I cannot remember who told me, but there somebody started playing with it already.
00:24:09.039 --> 00:24:17.039
Um making a mechanism that will you know try to look, I mean, based on instructions, go through the interface and then make screenshots.
00:24:17.279 --> 00:24:20.240
I'm not sure who told me that, but I think people are already playing with that.
00:24:20.720 --> 00:24:24.880
I've heard several different uh people have done it with different ways.
00:24:24.960 --> 00:24:31.839
Uh playwright, you you can use playwright to run um uh to record a screen and do it.
00:24:31.920 --> 00:24:43.680
I know Dulio had done some presentations on how he was able to build and take screenshots and build documentation uh and instruction manuals using page scripting, even when we did the page scripting session.
00:24:43.759 --> 00:24:53.519
He went through the example of how he was able to have an agent create a user's guide, including screenshots using playwright and that stuff.
00:24:53.599 --> 00:24:54.240
So interesting.
00:24:54.319 --> 00:24:55.599
Okay people are doing it.
00:24:55.759 --> 00:24:57.920
I I haven't done it for the screenshots yet.
00:24:58.079 --> 00:24:58.960
How about you?
00:24:59.440 --> 00:25:05.279
Me neither, but the question, the playwright, does it mean that somebody has to actually click on the interface?
00:25:06.240 --> 00:25:07.119
The human.
00:25:08.559 --> 00:25:10.720
No, based on page script, though, right?
00:25:11.200 --> 00:25:17.440
No, no, I I think you can do it autonomously to pull the information out.
00:25:17.599 --> 00:25:20.000
I haven't worked with it to do the screenshots, to be honest.
00:25:20.079 --> 00:25:26.319
I've worked with playwright only in the extent of page scripting, not in the from the respect of coding.
00:25:26.480 --> 00:25:32.000
But I know individuals are pulling screenshots, um, even in some of the groups that we're in, they talk about it.
00:25:32.160 --> 00:25:36.319
It's it's um yeah, yeah, documentation.
00:25:36.400 --> 00:25:45.200
It uh far better documentation than I've ever created, or any other developer that I've created, or any other consultant for that matter.
00:25:45.440 --> 00:25:53.119
It may not be 100% perfect, but if it comes back at 90% and it gives you a nice detailed document, and I have a few minor adjustments to make, I'm happy.
00:25:53.359 --> 00:25:54.000
Yes.
00:25:55.039 --> 00:26:04.400
Yeah, so I I definitely haven't I haven't uh had a contact with uh that yet, but I would love to know more about who's playing with it.
00:26:05.599 --> 00:26:13.759
There's so much going on with it, so that's uh uh yeah, it's uh it's an exciting time to say the least.
00:26:13.920 --> 00:26:19.039
It's I feel like I've never been busier than I have in my life.
00:26:19.200 --> 00:26:23.519
And sometimes I have to go back and say, what am I busy doing at some points?
00:26:23.599 --> 00:26:32.160
Because sometimes it's it's it's having fun, just you know, a lot of those things that I never were able to a lot of those things I was never able to do, I can now do.
00:26:32.480 --> 00:26:38.400
Where it would have taken weeks or months to complete, now I can get done, and sometimes hours or days.
00:26:38.960 --> 00:26:40.799
But do you feel you have more spare time?
00:26:41.039 --> 00:26:42.559
Because I don't.
00:26:44.240 --> 00:26:45.920
It's a misconception, right?
00:26:46.000 --> 00:26:50.079
Like everyone thinks like, oh, there's this AI supposed to do all the things for you.
00:26:50.240 --> 00:27:00.319
You I I think I think a lot of us that loves the product or just love building things, when you have a new tool, you always want to play with a new tool, right?
00:27:00.400 --> 00:27:02.319
So you always want to do something with a new tool.
00:27:02.400 --> 00:27:04.480
So you you you you you lose that time.
00:27:04.720 --> 00:27:13.119
I mean, just like um it's for example, when I when I got a motorcycle, I just want to get on a motorcycle all the time because it's like it's a new thing.
00:27:13.279 --> 00:27:16.960
Um, but yeah, no, you I don't have a lot of time.
00:27:18.240 --> 00:27:20.960
You want to play with this thing, you want to do more.
00:27:21.519 --> 00:27:22.319
So it's so fun.
00:27:22.960 --> 00:27:40.799
That is the I think the the common message from a lot of people that use it and get into it is any free time that I have, I would gain, I'm actually I'm able to now do the things that I was not able to do before.
00:27:41.359 --> 00:27:52.960
Because if you want to create a note-taking app, or if you want to create something uh else, you can now do it in a in a faster pace.
00:27:53.039 --> 00:28:01.359
So you can get more you have more time to do more things, and it's it's almost like feeding a fire.
00:28:01.519 --> 00:28:04.720
So it's to answer your question, no.
00:28:05.279 --> 00:28:08.880
I have to no, I do not.
00:28:11.440 --> 00:28:16.880
I have to remind myself to get up and and stretch and to eat and to do stuff sometimes.
00:28:17.119 --> 00:28:20.160
So yeah, yeah, I share the sentiment.
00:28:20.480 --> 00:28:25.119
It's good that you were in you working with it though, and uh you're doing some other things.
00:28:25.200 --> 00:28:27.680
You said you're working some Microsoft stuff as well, too.
00:28:27.759 --> 00:28:29.039
You're doing contributions.
00:28:29.680 --> 00:28:36.480
Yeah, so generally speaking, uh my work is dominant predominantly connected to the German market right now.
00:28:36.640 --> 00:28:45.599
I'm also I have done some fixes in in relation to the local formats being applied there for extra electronic exchange of invoices.
00:28:45.839 --> 00:28:49.119
It's called Reckon or Zugfair, if you have heard of it.
00:28:49.359 --> 00:28:51.200
If not, that's also fine.
00:28:51.519 --> 00:28:58.240
Um, I'm also uh you know that Microsoft is also uh changing a paradigm regarding the layout.
00:28:58.319 --> 00:29:10.079
I mean changing, let's say I don't want to have the hard cut, but we can clearly see that the reaction is set towards uh swapping the layout, the document layout, report layout.
00:29:10.559 --> 00:29:14.240
That's also a topic where I'm kind of uh familiar with.
00:29:14.799 --> 00:29:22.000
I'm wondering, for example, that area in to which extent can can you know AI help you with to create a new layout?
00:29:22.880 --> 00:29:33.200
And I think that's uh one of the areas which is not yet fully covered with this um AI workflow yet, but I think that might also change, I would say.
00:29:34.559 --> 00:29:35.519
I would think so.
00:29:35.599 --> 00:29:39.200
Uh so have you used AI in any of that at this point?
00:29:39.279 --> 00:29:40.720
Or yeah, yeah, yeah, yeah, yeah.
00:29:40.799 --> 00:29:41.119
I did.
00:29:41.519 --> 00:29:49.119
And my impression is it can already do RDL quite decently because RDL is already HTML formatted.
00:29:49.359 --> 00:29:55.119
I think it struggles more with uh with words, generally speaking.
00:29:55.359 --> 00:29:56.640
I mean it really depends.
00:29:57.039 --> 00:30:08.319
I mean when I say struggle, it can actually create really good looking layouts out of the box, but then the mapping is more trickier part, I would say.
00:30:08.559 --> 00:30:19.519
So I would say the forkflow is not 100% covered yet, because I think the AL development is ready to be covered, I would say, and this area is not yet covered, right?
00:30:19.680 --> 00:30:21.519
And I would definitely like to see it covered.
00:30:21.759 --> 00:30:29.839
Imagine that, like you are saying to the business central your topic here is the check, here is how it looks like.
00:30:30.240 --> 00:30:38.559
Build me a word layout, and you know, map me the fields in there and take care that the dimensions are important, they should be precise.
00:30:38.880 --> 00:30:46.079
That's you know what you know, in the ideal world, what would AI should bring us as a quality?
00:30:46.640 --> 00:30:48.559
We're not there yet, that's for sure.
00:30:49.200 --> 00:30:53.839
But yeah, some of the task that I'm uh thinking about right now.
00:30:54.319 --> 00:31:14.319
If it can do a US check, which I will always say I still don't understand why we use checks, but if it can do a US check, that would be rather impressive because for some reason I s after doing this for decades, I still don't understand why checks are so troubling.
00:31:14.559 --> 00:31:20.799
And it's not business central, it's just replicating the check, I think, for some reason.
00:31:21.039 --> 00:31:25.599
And sometimes people just get stuck on having a particular check format.
00:31:25.839 --> 00:31:27.359
I don't understand it versus me.
00:31:27.440 --> 00:31:30.799
I would say, okay, I can buy stock, this check fits it, okay, done.
00:31:31.119 --> 00:31:33.359
But some people like to have yeah, but we don't.
00:31:33.440 --> 00:31:34.480
I mean, let's forget the check.
00:31:34.559 --> 00:31:42.880
Let's say that in any country or in any partner there might be some customer-specific uh report or layout that they want to print out.
00:31:43.039 --> 00:31:44.799
I mean, labels, whatever.
00:31:45.119 --> 00:31:46.880
And that's what I dream of.
00:31:46.960 --> 00:31:48.160
Like here is the prompt.
00:31:48.880 --> 00:31:53.359
I may give you something that looks appealing to me and bring it on.
00:31:53.519 --> 00:31:56.240
Right, here is the data set design everything.
00:31:56.400 --> 00:32:04.960
Because I think I I think a lot of the time partners charge today is at least in Germany, is related to the reports themselves.
00:32:05.039 --> 00:32:07.440
I mean, and not only in Germany, but like all over the world.
00:32:08.079 --> 00:32:13.440
And if we could, you know, get that out of the way, that would be the next step forward, I would say.
00:32:14.720 --> 00:32:15.279
I agree with you.
00:32:15.440 --> 00:32:17.359
Yeah, checks are I mean, not checks, yeah.
00:32:18.400 --> 00:32:27.680
Reports are a big part of some of the implementations, and having a way to make that easier.
00:32:27.920 --> 00:32:41.119
Uh custom report layouts was one step, then adding the word add-in was another step to making the transition to having report and and well, even back up doing report extensions to some extent.
00:32:41.279 --> 00:32:51.759
All these things helped make the reporting easier, but I think seeing it where you had mentioned that AI can now work with the Word document.
00:32:51.920 --> 00:32:58.000
I haven't experienced it, so if it can do it, or if anyone else that's listening has done it, I'd uh we'd like to hear about it.
00:32:58.160 --> 00:32:59.440
But I think that would be great.
00:32:59.519 --> 00:33:11.440
Or I was even just thinking as you were speaking, even having an RDLC to Word, I don't even know if it'd be possible, but conversion or some type of uh setup would also be great.
00:33:11.759 --> 00:33:12.960
Yeah, so I tried it out.
00:33:13.039 --> 00:33:19.599
It's I would rate it like I don't know, 60-70% success rate.
00:33:19.759 --> 00:33:22.240
So you still have to fill it a bit.
00:33:22.640 --> 00:33:24.480
I mean not a bit, but more than a bit.
00:33:24.640 --> 00:33:28.000
Uh so sometimes it produces the RDL which cannot be compiled.
00:33:28.160 --> 00:33:32.160
Uh the mapping in Warlayout I haven't succeeded yet.
00:33:33.440 --> 00:33:37.599
But still, I would still apply it and then like finish up the rest.
00:33:38.160 --> 00:33:43.920
So at this point of time, I'm just saying we haven't um made the complete journey yet.
00:33:44.640 --> 00:34:02.640
So you haven't gone the full path with it, but it still helped out enough to where it was a good start, where it wasn't sometimes you go through these exercises and you say, Wow, it's so poor that I have to spend more time fixing it than if I were to just do it myself.
00:34:02.880 --> 00:34:07.119
But you're saying that from your experience, it's a good starting point.
00:34:07.680 --> 00:34:08.960
Yeah, that's my opinion.
00:34:09.440 --> 00:34:10.000
Yeah, yeah.
00:34:10.559 --> 00:34:15.679
I mean, for for start, uh I I don't know, Claude can already chew up.
00:34:16.320 --> 00:34:21.039
I mean, the the prompt that I gave him is like here is the invoice from the business central.
00:34:21.599 --> 00:34:25.360
How it looks like it's a bit uh how they say old fashioned.
00:34:25.920 --> 00:34:29.679
I want appealing invoice in the war layout, and it's fitted out really good.
00:34:29.920 --> 00:34:35.519
So the cloud can already do quite a lot with PowerPoint, Word, and so on.
00:34:35.840 --> 00:34:40.320
I the missing link is you know connect it properly.
00:34:43.679 --> 00:34:44.960
Now my mind's coming.
00:34:45.039 --> 00:34:46.960
Well, I I'm sure that you'll be able to get that.
00:34:47.039 --> 00:34:55.679
It's a good journey for you to to explore, and I'd be interested to hear the results of how far you get with it as the technology increases as well.
00:34:55.920 --> 00:35:00.639
Yeah, but you're you're what you're one of the few that loves to do reports.
00:35:02.000 --> 00:35:04.400
Uh I wouldn't phrase it that way.
00:35:11.760 --> 00:35:14.719
Yeah, I don't know whether we can make that statement for anybody.
00:35:14.880 --> 00:35:19.760
It's just, you know, some topics land on you at some point in time and you want to get them done.
00:35:20.320 --> 00:35:24.400
I'm sure I mean some people are more successful on avoiding the topics.
00:35:24.559 --> 00:35:25.360
I'm I'm not.
00:35:25.519 --> 00:35:27.840
So this is in front of you.
00:35:28.000 --> 00:35:28.719
Can I get it done?
00:35:29.280 --> 00:35:34.880
I won't say it too loud, but I've been very successful at avoiding that topic for some years.
00:35:36.480 --> 00:35:38.960
Someone's gonna ask you, Brad, the next time you go speaking.
00:35:39.280 --> 00:35:40.480
Good for you, good for you, good for you.
00:35:42.159 --> 00:35:50.559
No, I've done I've done, I can count how many reports I've created on one hand in the past year or two, maybe two, less than two hands.
00:35:50.719 --> 00:35:52.480
Uh yeah, maybe one hand.
00:35:52.719 --> 00:35:53.519
Uh thankfully.
00:35:54.079 --> 00:36:00.159
If you do everything right, you probably will not have to touch them, you know, because you'll just say, give me what I need and that's it.
00:36:00.639 --> 00:36:03.679
Yeah, that's uh that'll that'll be nice.
00:36:03.920 --> 00:36:18.559
Um I've always primarily focused more on uh the data portion of it, you know, being able to have the data that someone else can either write the report or they can use the information somewhere else to report and analyze their data.
00:36:18.800 --> 00:36:29.360
So uh I'm not if I'm not touching those, that project to get AI to do uh word layouts, but I think it would be a good one to see.
00:36:29.440 --> 00:36:40.480
Uh and I know again, as you had mentioned, that Microsoft's making a push to replace a lot of the standard customer-facing documents, as we'll call them, from RDLC layout to word layout.
00:36:40.559 --> 00:36:46.719
Uh, from my understanding, there isn't a date that says RDLC won't work.
00:36:46.800 --> 00:36:51.840
Uh anybody that has RDLC reports will still be able to create those, they'll still be able to work with those.
00:36:52.000 --> 00:37:00.960
Uh, I'm sure as as technology advances, though, RDLC will fall off because uh it's my understanding that they're not doing a lot to improve it as well.
00:37:01.280 --> 00:37:04.719
Uh it's actually not even the business central topic at all, right?
00:37:04.800 --> 00:37:07.280
So it's not the business central team that's maintaining it.
00:37:07.360 --> 00:37:10.800
So I would say it's probably uh a bit out of their hand.
00:37:11.039 --> 00:37:13.840
I mean, I I think they will support it as long as they can.
00:37:14.159 --> 00:37:14.480
Yes.
00:37:14.719 --> 00:37:17.599
Because they don't want to lose customers because of that.
00:37:18.079 --> 00:37:21.760
But like I I think they are not controlling the plug.
00:37:23.840 --> 00:37:25.039
No, I mean I don't think so.
00:37:26.880 --> 00:37:27.760
Yeah, yeah.
00:37:29.039 --> 00:37:33.199
Yeah, and the other topics I'm dealing with right now are the performance, for example.
00:37:33.280 --> 00:37:49.199
Um stuff, and I think um uh I I have looked at the stuff that Torben did with this um uh website web to app up that can analyze your profile profilers and it's something also which I see potential, right?
00:37:49.280 --> 00:38:04.800
Uh imagine yourself uh solving the performance problem by giving you know the food to some of the tools which will tell you what to improve right away without you having to think that is a big place.
00:38:05.199 --> 00:38:26.159
I know there's there's many initiatives now that are all working towards having AI analyze code, analyze telemetry to be able to help identify non-performant code and and ways to improve the code, which I think is extremely beneficial.
00:38:26.480 --> 00:38:44.719
I support all those initiatives because I think that Business Central is a great application, and you know, maybe because I've been using it for decades now and I've been in it for decades, and it's it's how I've made my career, basically.
00:38:45.679 --> 00:38:58.719
I want successful implementations, and if you have non-performant code at an implementation, it can make a customer extremely unhappy with Business Central, and it's not even the Business Central application that's the problem.
00:38:59.440 --> 00:39:21.280
So having tools that allow developers from anywhere, from ISVs, partners, customers, anybody who's developing with Business Central, having the tools to analyze the code to be able to help identify performance issues helps the product because customers will be happier with the performance of the product.
00:39:21.599 --> 00:39:30.320
Because I think we've all heard this is slow, this takes too long, and all those other things over the years.
00:39:30.480 --> 00:39:41.280
So it's um it's great to have tools now or in in the works to be able to help uh mitigate some of those uh performance bottlenecks.
00:39:42.239 --> 00:39:42.719
Exactly.
00:39:42.880 --> 00:39:45.199
I've yet, for example, to to explore.
00:39:45.360 --> 00:39:49.679
I'm not uh telemetry prof you, not sure how do you did you deal with telemetry so far?
00:39:50.480 --> 00:39:53.280
I've worked with uh I worked with a lot of telemetry.
00:39:53.360 --> 00:40:01.920
Uh I did not ex I'm it's on my list now to work with Waldo's BC telemetry tool because we have a lot of telemetry.
00:40:02.079 --> 00:40:05.599
Uh unfortunately, I have a lot of things on my list to do.
00:40:05.920 --> 00:40:09.360
And that list AI is supposed to be making better, Brad.
00:40:09.440 --> 00:40:11.119
Come on, it's supposed to be faster.
00:40:11.840 --> 00:40:16.000
You still have yeah, it's it's I'm just messing with you.
00:40:16.239 --> 00:40:17.039
No, I know you are.
00:40:17.280 --> 00:40:20.639
It's it's um it's it's one of the next things that I want to look at.
00:40:20.719 --> 00:40:24.639
And then it's also a topic where I'm also falling behind.
00:40:24.719 --> 00:40:25.920
That's my feeling right now.
00:40:26.079 --> 00:40:26.320
Yeah.
00:40:26.880 --> 00:40:27.599
So yeah.
00:40:27.920 --> 00:40:41.199
I I've been using a I'm I'm used to doing a lot of KQL and stuff like that, but from what I understand, to not have to, you know, to be able to read the telemetry logs uh through the MCP is uh exciting.
00:40:41.280 --> 00:40:45.360
So I'll probably set that up next week is my plan.
00:40:45.679 --> 00:40:53.039
Uh because I'm going to bring back an initiative that we have to um better utilize telemetry.
00:40:53.840 --> 00:40:55.599
Yeah, I think that's yeah.
00:40:55.840 --> 00:40:57.920
I think the whole community actually should work.
00:40:58.079 --> 00:41:00.480
I mean, improve on that area, generally speaking.
00:41:00.639 --> 00:41:09.440
My feeling is actually not that many people are are not that many partners are really onto that, but they should yeah.
00:41:10.000 --> 00:41:10.639
100%.
00:41:11.280 --> 00:41:18.960
100% because the telemetry you can identify many more I don't want to say problems, but you can I will say problems.
00:41:19.039 --> 00:41:28.559
You can identify many issues quickly within telemetry, uh with identifying code that's non-performant or identifying other issues.
00:41:28.719 --> 00:41:37.519
It's uh either even issues outside of code, I think you can identify because you can capture errors that the users are making as well, too.
00:41:37.840 --> 00:41:46.239
So for telemetry, I mean it you know, I haven't really played with telemetry at all in Business Central, but you know, uh in passing.
00:41:46.400 --> 00:42:07.679
So if you if you were to work on telemetry and having an agent read the telemetry data, you could you could probably perhaps take an agent and look at um uh maybe some patterns that are uh you know kind of weird uh or it's something that's out of whack and create issues um out of it as well.
00:42:07.760 --> 00:42:11.760
So then you can immediately take action against it.
00:42:12.000 --> 00:42:19.920
Um maybe create automatically create GitHub issues and say, hey, look, based on telemetry, we see these weird patterns that's happening around this time.
00:42:20.159 --> 00:42:27.519
Uh rather, right now, I mean I'm assuming everyone's gonna look at looks at the telemetry and says, Oh, okay, I think maybe that's an issue.
00:42:27.679 --> 00:42:30.000
It's it's sort of a manual process right now.
00:42:30.159 --> 00:42:32.079
I think what's everyone doing.
00:42:34.000 --> 00:42:34.320
Yes.
00:42:34.480 --> 00:42:34.800
Yeah.
00:42:34.960 --> 00:42:36.960
No, it's it's you can use it to be proactive.
00:42:37.119 --> 00:42:50.079
Telemetry was always one of those things you could use to be proactive if you had the right dashboards, if you had the right queries set up that you could run and monitor, but with AI and then utilizing some of these tools, you could absolutely again be proactive.
00:42:50.159 --> 00:42:52.239
So you could identify issues.
00:42:52.559 --> 00:42:54.000
And I use the word issues loosely.
00:42:54.079 --> 00:42:59.920
You could identify you could see code that needs to be refactored, you could see other uh things with telemetry.
00:43:00.159 --> 00:43:13.039
And I've kind of tried to list all the things you can identify that you can, again, as you had mentioned, maybe if you make an issue for someone to take a look at or even AI take a look at within the code, will help make the customer have a better implementation.
00:43:13.199 --> 00:43:20.480
And when I say customer, it's not just partners, not just ISVs, it's even customers that have their own internal development team, they can take advantage of these tools as well.
00:43:20.719 --> 00:43:23.760
So it's it's anybody that's in that space.
00:43:24.000 --> 00:43:27.119
Um it's it's tough to keep up with all this though.
00:43:27.599 --> 00:43:28.320
Yeah, true.
00:43:28.880 --> 00:43:35.920
This is this is why we don't sleep, because we have to keep up with not only AI, but all the features in Business Central that keep changing as well.
00:43:36.239 --> 00:43:41.039
So it's uh it's it's challenging, I I think.
00:43:41.760 --> 00:43:42.400
Yeah.
00:43:43.679 --> 00:43:46.239
Are you are you guys doing anything outside?
00:43:46.320 --> 00:43:54.320
Like I know with all this AI, there's so much to do in Business Central with uh with some of the AI co pilot using co pilot cloud and things like that.
00:43:54.400 --> 00:43:56.800
Are you guys doing anything outside of that?
00:43:57.039 --> 00:43:59.760
I mean, I'm not trying to add more things to do.
00:44:00.320 --> 00:44:04.159
For you, but are you doing anything outside of Business Central?
00:44:05.440 --> 00:44:09.199
Uh well, I know I'm I mean with or without AI.
00:44:09.280 --> 00:44:16.960
I mean I I have I have personally I use I use uh GitHub compiler, I use git uh cloud code, uh but also cloud web app.
00:44:17.840 --> 00:44:23.119
Uh it writes me everything actually, whenever I want to phrase anything, it's basically no longer.
00:44:23.280 --> 00:44:27.199
I mean if it's in text, it doesn't go without AI review.
00:44:27.360 --> 00:44:35.840
I also have Gemini aside because I have a phone where where I mean there is some consumption promotion and you get the Gemini automatically.
00:44:37.119 --> 00:44:42.000
And yeah, currently I use them in parallel depending on the needs.
00:44:42.320 --> 00:44:48.239
I guess most of the people who try, I mean, they try to use what works the best at the point of time.
00:44:48.719 --> 00:44:55.519
I mean the paradox is actually yeah, do you have like three or four subscription different subscriptions?
00:44:58.480 --> 00:45:05.280
I use Claude and GitHub Copilot, uh the two uh frameworks that I use.
00:45:05.679 --> 00:45:13.119
And the models that I use, I primarily use Opus and Sonnet at this point for my tasks.
00:45:13.440 --> 00:45:16.559
Uh the other ones I haven't used in quite a while.
00:45:16.719 --> 00:45:20.400
Uh so yeah, I do have those are the subscriptions I use.
00:45:20.480 --> 00:45:21.039
It's tough to try.
00:45:21.280 --> 00:45:23.280
I you you know, I think I just stuck with those.
00:45:23.440 --> 00:45:24.880
Uh I tried them out.
00:45:25.119 --> 00:45:27.360
And I I I you I'm like you.
00:45:27.440 --> 00:45:32.079
I use Claude, the web app, I use Claude Code uh to do everything.
00:45:32.800 --> 00:45:35.599
It's it's i I I have it do everything.
00:45:35.760 --> 00:45:39.280
It's yeah, I was looking at uh the to move.
00:45:39.360 --> 00:45:46.480
So when I was looking at um properties, uh I'd feed them the listings, it'd come up with this nice checklist, this nice chart.
00:45:46.639 --> 00:45:47.119
It does everything.
00:45:47.199 --> 00:45:47.679
It's amazing.
00:45:47.760 --> 00:46:00.239
It gives you like a nice portfolio, it can do research, and then as you go through the properties, uh you tell it you're at a property, it will start asking you questions, you can tell it what you want, like even with voice, because on the phone you can record and go voice to text and you can use voice.
00:46:00.400 --> 00:46:03.280
And then it just continually adds this nice document portfolio.
00:46:03.519 --> 00:46:06.800
It's AI is so much more than development.
00:46:07.199 --> 00:46:16.639
So you're doing fun stuff outside uh uh fun are you you're building things outside of just business central, it sounds like you're doing personal projects.
00:46:16.880 --> 00:46:18.159
Absolutely, I am.
00:46:18.400 --> 00:46:20.000
Are you as well?
00:46:20.239 --> 00:46:25.760
Yeah, for myself, I I don't do some business central is is the main focus for me.
00:46:25.920 --> 00:46:30.480
I mean, uh but uh what I would like to try is uh claw, right?
00:46:30.639 --> 00:46:33.760
Setting up something at home, doing some stuff.
00:46:34.000 --> 00:46:36.320
I know a guy that will help me out.
00:46:38.480 --> 00:46:49.599
Uh yeah, but uh but other than that, let's say I have used it, uh you know, I have been on some sessions recently, and I'm going to some more, and I use it for PowerPoint generally speaking.
00:46:49.840 --> 00:46:51.599
Yeah, yeah.
00:46:51.679 --> 00:47:04.639
And I think it's all basically depends on which area of of I mean what's your um focus on I would say in in life and you probably can find an application how yeah I can would help you.
00:47:05.280 --> 00:47:07.599
Probably at least that's for me, like that.
00:47:07.679 --> 00:47:14.719
So I mean, even my kids started criticizing me that I mean they already called me Chat GPT, although I don't use it any longer that much.
00:47:16.159 --> 00:47:17.679
They make fun out of me.
00:47:19.519 --> 00:47:21.360
And it's also a funny effect, you know.
00:47:21.440 --> 00:47:33.760
I have also heard of a podcast about the fact that you know, people when you use AI that much, you kind of stop thinking yourself at some point of time.
00:47:34.000 --> 00:47:38.719
So it's true can be contraproductive when you use AI too, let's say.
00:47:38.960 --> 00:47:51.039
And that's also something that's not widely spoken about, but basically that's also a side effect of I mean uh uh less stimulation for our brain is when you let them think everything.
00:47:51.679 --> 00:47:53.440
That's yeah, a very good point.
00:47:53.599 --> 00:47:54.800
It's like the calculator.
00:47:55.039 --> 00:48:03.119
It's when I was younger, when I was using an abacus, and I had to do math in my head, and yeah, I could do math really well.
00:48:03.199 --> 00:48:11.199
And now, because of the computer and the calculators, uh, you know, ask me what you know 10 divided by one is, and I have to think for a moment.
00:48:11.519 --> 00:48:30.239
It's but you can do other things uh to to hopefully stimulate your your mind to to give it uh uh some some excitement because as you said, if you just let AI do everything for you, then uh there's no fun in that.
00:48:30.480 --> 00:48:41.039
It's fun having some thoughts and and uh uh creating and architecting and designing, but uh you know, the way we're developing and creating outputs going to change.
00:48:41.199 --> 00:48:44.159
Just imagine you'll have to tell your grandkids or something.
00:48:44.320 --> 00:48:47.199
I remember back when I was developing, we had to write code.
00:48:48.800 --> 00:48:50.079
Just think about that.
00:48:50.880 --> 00:48:52.000
Yeah, yeah, yeah.
00:48:52.800 --> 00:48:54.719
I mean, it's gonna be even more funny.
00:48:54.800 --> 00:49:11.119
I think you're a spot on because I have heard uh you know, interesting statement from Jeremy, and he said, Yeah, I mean, that's one probably he I mean that's already being uh discussed about I mean the the the languages that we use today for development they are built for humans, right?
00:49:12.559 --> 00:49:28.000
But we'll be seeing in in the following years the languages that are developed for AI that we will actually not be able to read for them at all because we can read the the languages that are built for us, but they are probably not the most optimal for AI.
00:49:28.960 --> 00:49:31.119
We'll be seeing so anyway.
00:49:31.519 --> 00:49:32.480
I think I think so.
00:49:32.639 --> 00:49:42.320
I think we'll go back because following on that conversation yes, I took assembly language in college too.
00:49:42.719 --> 00:50:01.119
But you you think back with assembly language was written directly for the computers, and then you had IDEs and compilers that translated that information and created executables, and it was just easier for people to program instead of remembering doing pop 10, push, all those other things that you had to do.
00:50:01.360 --> 00:50:02.960
I think you will have that point.
00:50:03.119 --> 00:50:09.840
And someone even I saw something or I had a conversation with someone said, No one checks the executables.
00:50:10.000 --> 00:50:16.559
We all assume that the compiler is going to build the executable properly based on the code that it received.
00:50:16.800 --> 00:50:20.000
I do think we'll get to a point, we're in this transition period.
00:50:20.239 --> 00:50:29.519
I do think we will get to that point where the human language will be what you speak to a tool, and that tool will create the executable that does what you want.
00:50:29.760 --> 00:50:33.920
Right now, we're thinking like humans, right?
00:50:34.000 --> 00:50:47.679
We're assuming that the AI is going to create the human text and the human literature, but I think it will get to the point where it will be able to um work with just what it needs to take over the world.
00:50:48.079 --> 00:51:00.480
I thought that I thought I remember reading an article I read around that where uh they had two agents uh to be more efficient communicating to each other, they created their own language.
00:51:00.880 --> 00:51:08.960
I'm pretty sure there's an article that I read about that, and I thought it was pretty fascinating that it created its own language because it was not efficient enough.
00:51:09.199 --> 00:51:17.199
So it decided and said, okay, for us to communicate together, let's create a language where we can be uh more efficient in how we communicate.
00:51:17.519 --> 00:51:25.119
And uh I'll have to find that article, but I thought it was pretty fascinating that eventually it'll come to a point that that's what it'll do.
00:51:26.400 --> 00:51:36.880
Yeah, that will be that'll be an interesting time, and then you know, then we'll have the world.
00:51:37.440 --> 00:51:38.639
I I I don't even know.
00:51:38.719 --> 00:51:40.880
Like all those old movies that we've had.
00:51:41.119 --> 00:51:44.719
But uh no, thank you very much for taking the time to speak with us this morning.
00:51:44.800 --> 00:51:45.599
It was a great conversation.
00:51:45.679 --> 00:51:47.119
I could talk about this stuff for days.
00:51:47.199 --> 00:51:49.039
We really do appreciate you taking the time.
00:51:49.199 --> 00:51:58.320
If anyone would like to contact you to talk a little bit more about AI, reports, word layouts, how to get them done, or anything else that you're doing, what's the best way to get in contact with you?
00:51:58.639 --> 00:51:59.760
Um, I have a blog.
00:51:59.920 --> 00:52:07.039
Just go to my blog, uh, to the contact form, or send me an email uh on uh millionrs at househood.com.
00:52:07.840 --> 00:52:10.480
Excellent, great great, thank you.
00:52:10.639 --> 00:52:12.159
And uh look forward to seeing you again soon.
00:52:12.320 --> 00:52:14.079
It was great meeting you last week in person.
00:52:14.239 --> 00:52:27.840
It's great, you know, we see all these individuals within the business central community from all parts of the world, and you have communications with them, you see some of the stuff they're doing, and it's always nice to finally get to meet people in person.
00:52:28.000 --> 00:52:29.360
So uh thank you guys.
00:52:29.440 --> 00:52:30.079
It was fun.
00:52:30.400 --> 00:52:34.320
I think it's always fun talking to you, and thank you for what you're doing for the community.
00:52:34.400 --> 00:52:39.519
I think it's really putting some value to everybody uh interested in this topic.
00:52:39.840 --> 00:52:40.960
All right, great, thank you very much.
00:52:41.119 --> 00:52:43.119
Have a good afternoon and talk to you soon, good afternoon.
00:52:43.360 --> 00:52:43.760
Bye bye.
00:52:44.000 --> 00:52:44.559
Bye.
00:52:44.880 --> 00:52:52.159
Thank you, Chris, for your time for another episode of In the Dynamics Corner Chair, and thank you to our guests for participating.
00:52:52.400 --> 00:52:54.000
Thank you, Brad, for your time.
00:52:54.159 --> 00:52:57.679
It is a wonderful episode of Dynamics Corner Chair.
00:52:57.840 --> 00:53:01.199
I would also like to thank our guests for joining us.
00:53:01.440 --> 00:53:04.159
Thank you for all of our listeners tuning in as well.
00:53:04.400 --> 00:53:08.239
You can find Brad at developerlife.com.
00:53:08.400 --> 00:53:12.800
That is D V L P R L I F E dot com.
00:53:13.039 --> 00:53:18.559
And you can interact with them via Twitter, D V L P R L I F E.
00:53:19.199 --> 00:53:27.440
You can also find me at mattalino.io, m-a-t a l i no dot io.
00:53:28.239 --> 00:53:31.920
And my Twitter handle is mattalino16.
00:53:32.719 --> 00:53:35.679
And see you can see those links down below in the show notes.
00:53:35.840 --> 00:53:38.960
Again, thank you everyone, thank you, and take care.