Episode 515: Stop Coding, Start Architecting: How AI Is Reshaping the BC Developer Role

Cold Open On Parallel Tasking

Task worker. It's like a library for developers where you can able to uh easily to split some uh huge slow processes into chunks and run it in separate sessions. Because I encounter this problem very often in my experience.

Welcome everyone to another episode of Dynamics Corner. What does Two Cats have anything to do with AI, workflows, framework, etc.? We're about to find out. I'm your co-host, Chris.

And this is Brad. This episode is recorded on April 10th, 2026. Chris, Chris, Chris. Two cats in AL development. That's an interesting concept. Also, we learned about tips and tricks for starting your AI AL development journey, as well as what you can do with your free time now that we're using in living in the world of AI. With us today, we had the opportunity to speak with you doing.

It's morning for me. It's eight o'clock in the morning right now.

Oh, it's your also morning.

Yeah.

I see.

It's uh it's 6 p.m. for me, so it's for me, it's evening.

Oh dinner, dinner time. Your day's over, it's good. I like uh I like the the cats that you have joining you as well.

Um it's not planet. Immediately when I uh join uh meeting during our session, they just decide to be here. I think they would like to.

They wanted to join with you. What are their names?

So uh white cat is uh he is uh Hades.

Say that again, please.

Hades. It's like uh yeah, Hades, Hades. From the Greek uh Yeah, from Greek mythology, it's actually yeah. And she is Nyx.

What is that again?

Nix Nix, it's also from Greece uh mythology as well. So that's like some interesting uh name choice, but uh I I love them, they I like them, and uh they're like very new for me. Uh only two years ago.

Business Central 2026 Wave 1

I I I I it's it's nice, it's it's good to have a companion uh with you when you're you're sitting at home working, and it's it's been a wild, crazy couple of months with uh with technology, with development, with AL. Business Center released Wave 1 for 2026, just a short period ago. Usually, and I have to figure out why, usually I go through all of the features well before the release and experiment and play with them. But I think with the way things are going, I've been extremely busy and and I was doing some reflection and figuring out why am I so busy. And I think part of it is because in the age of AI and development, we can do so much more, so we're busier doing more, whereas before we would take the time to do things. And I was looking at the list of development editions for 2026 Wave One, and I'm curious to know if you had the chance to work with them because I because I just started working with them recently. Uh some of them I'm really excited about.

I actually take um short look on it. Uh I also was not able to handle it uh fully, and uh it's like it was a lot of uh new changes regarding agents, if you notice. Uh like now you can uh build some agents in uh your sandboxes environments, and uh after that you can export it and use it in code. Uh in theory, it's just uh perfect idea. I like it. Uh there's many different changes uh regarding um how uh you can use unified interfaces to work with IA inside business central, like um those specific um code units interfaces. And I even use it, I I I try to use them. But some of some of my part of experience was great, some of it's not great, like there are some limitations, like you cannot uh pass um files and uh work with them efficiently, like it's it's still in uh in the stage where it's evolving, and there are some challenges which we are uh right now trying to uh to solve and Microsoft trying to solve.

So yeah, it's yes it it is, it's exciting. So it's the trouble I haven't played with it. I I've looked at some chats, but the enabled troubleshooting MCP server for AL, I really want to play with. I I know that's due in May, so I'm I'm looking forward to seeing that. The AL tests from Visual Studio Code, I'm excited about that um should be out now. And also, as you had mentioned, BC Bench or uh agent preview or agent playground to be able to create agents in the sandbox. I've created many of those and have been running the tasks through the sandbox user interface. I really do hope that they make it to where you can create the agents and not have to create the code for the agents with what you create with the preview.

But it feels it feels so natural to just not to quote it, but just to um like configure it and use it. I totally agree on that.

Yes, I I've created many and you go back and the task run, and I just feel like I should be able to click schedule task or run task or do something, and uh that's my hopes that one day we get there because as you had mentioned, it feels so natural, and you can do some uh great things. I've had some demonstrations that I've done with it, and that's what everybody's saying is why do we have to code it now? It's already there.

Uh maybe, maybe it won't take too long. Maybe uh give it a few months.

Yeah, it just it's just very fresh. Or maybe we can contribute somehow. Who knows? Maybe, but still in future.

That's what they want though. That's what they want, right? They want you to uh contribute and enhance it. Here's the tool. Make something right.

Volodymyr’s Background And Community Work

So so those are some of the big ones that we're excited about, but uh we definitely wanted to speak with you about a lot of things from business central uh with what you're doing from the technical point of view, because uh I know you're doing quite a bit in development and development workflow. Uh, but before we talk about that, can you tell us a little bit about yourself?

Uh sure. So um I'm Volodymyr, uh ABC developer from Ukraine, a recent Microsoft MVP, uh, and I have been in the ecosystem for nine years. Um I have worked with BC in different capacities. I work at B Dynamics with Swiss partner, Microsoft Partner, where I focus on enterprise level projects. Additionally, I run company Volodiner LC where I do freelance work for small medium-sized clients and publish my own app source extensions. I spent also a good chunk of my time contributing on open source and writing a technical blog at uh vldbc.com. Uh what I really focus on is more developer tooling, building things that make everyday life easier for BC developers, uh partners, customers. Um my open source uh product, data unitor tool, for example, uh, has over 160 stars on GitHub and almost 1000 installs on App Source, and uncounted number of installs on PT. I don't know, I didn't count it. And I enjoy Business Central. I do all of this because I believe in BC as a product and its community. I think open source helps everyone grow, and I love sharing what I learned, whether that's thoughts on my blog, at events, or just in conversations like this one.

That's great. So you you have the open source project that, as you mentioned, has a number of stars and a lot of installs. And I'd like to talk about open source projects, but some of these tools that you've created for development, how are they impacting the development workflow and the benefits to users of the tool when their development process?

Great question. Uh I think uh the idea behind it is to uh fill all these gaps what Microsoft are uh have because they are have many many focuses, like IA, uh all of these important uh tools as well. But there are some gaps which I can see uh that I can add some value and add some value to community to to create something that can be useful specifically uh to solve some specific problem. For example, data you tool trying to solve problems, uh how you can manage, edit, view, data mostly for testing uh in environments, as you remember in old days uh you were able to do it in Object Designer, uh in Navision Classic or RTC, doesn't matter, or even a SQL. Uh but after BC14, if I remember correctly, they change it and you cannot manage data. Uh same story. Uh recently I released um parallel task worker. It's like a library for developers uh where you can able to uh easily to split some uh huge slow processes into chunks and run it in separate sessions. Because I encounter this problem very often in my experience that some huge processes which you cannot make it faster, it's like impossible. There is no uh room for that. So I think uh idea behind it is to find some problem and trying to solve it. And if you don't see any solutions in um in the internet and community, then it's even best uh or greater uh idea to put some effort on it and to make life uh for developers and partners easier. That's what I think about it.

Open Source Tools That Close Gaps

No, no, it's important to see those tools. And you had mentioned, I know with Business Central, you can now, and you have been for many years, able to make contributions to the application to enhance the features and the functionality of the application. And then also, as you had mentioned, many individuals are making open source contributions to help with uh PTEs uh or for the application or even some development tools. I know a lot of individuals are adding development tools um uh uh options available. The cats are going a little crazy there. It's kind of why I had a brief pause. It's fighting, fighting. Yeah, I thought there was going to be a battle, a battle soon with uh with the cats.

They started when you mentioned open source uh contributions. Interesting. Why?

How AI Changes AL Engineering

They're excited about the open source contributions and well with that. Uh we appreciate all that you're adding, and you have a lot of great tools available on your uh GitHub repository. But another thing that we were talking about previously is the impact of AI in development in general, and in general also with AL. So you've been working with development for nine years, so you've come into the tail end, I believe, of um the old way, as everybody talks about with Cal, transitioned into AL, working with AL. And now, as we talked about, Microsoft's even adding development tools, as we talked about the introduction to this session. How has AI impacted your AL development, or how do you see it impacting AL developers?

That's a very uh great question. Let me think about it more. Um first of all, I would like to admit you mentioned very correctly there is not only open source projects which you maintain by your own, there's also contributions to uh direct contributions to business central uh itself. You can uh contribute uh as open source contribution uh to system application, to uh base application also as well. And uh I would like to like to highlight it to all developers uh who listen it, or maybe will listen it, don't afraid to contribute to business central itself. It's actually add a lot of value, it's uh you solve your problem, your community problem, and product itself will gain some value from it. So don't afraid don't afraid to do it. Start with something simple. And uh back to your question uh about the effect of uh how artificial intelligence can affect uh development process specifically for AI and in general. Um I will start with uh some valuable uh insights. What I found out when I uh start to um realize that I is closer and closer uh in development process. I start to use uh IA firstly, uh I think when uh GPT 3.5 was released. Uh it is 2020, 21.

Not I don't remember exactly the year, but it goes so fast I can't keep up with it either. It seems like yesterday a lot of times that something was released and someone will tell me it was months ago or years ago.

AI Testing And TDD Become Standard

I totally agree, it's evolving very fast every month, every six months, every uh place or situation is changing dramatically. That's what's happening. Uh and today, if you speak about today, uh what I actually see that uh IA is to make code is cheaper, you can generate and create many uh many products uh without spending much amount of time uh for writing all specific line of code. You can spend more amount of time uh to focus on architecture of product, how you can build things, you can do actually real software engineering, because if you think about software engineering, it's not just about uh uh write some lines in text in on in in in notebook, it's actually to design, to think about architecture, to think about limitations, to think how it will be maintained in the future. And I think II give us a huge uh tool to avoid some um boring work which are not super actually software engineering if you think it about more. So I think it gives you an opportunity to generate a lot of code, but at the same time, I think uh it's also a dangerous um because when you will generate too much line of code, how you will understand them very well. Someone will tell you don't need to understand it because you can just uh delegate all work to agent and it will do everything for you. Maybe, but we are not on this stage where we can avoid reading code, we still need to read code, uh, and there is no real huge product uh in the market which was fully wipe coded without reading and fixing the real code. It's still a case where we need to read code. And when we generate too many lines of code, we also uh lose some um how to say, you are you are not fully understand what it was, uh how it was generated, how it actually works, because you are not who writes these lines, you are not aware of them, and you have no psychology bound to project. You are not understanding it uh so good as you can do it when you write it. So I think agents can also create problems here. So there is some performance speed up, and there are some issues with quality, I think. It's my from my my perspective, uh it can be some issues with quality, but still many of these issues can be fixed with proper usage of it. Like you can uh generate uh you can use some strict uh uh limiters uh to avoid some standard uh problems, you can improve uh what uh tools you use inside your agent, for example, uh you can improve uh system, how you work with uh cloud or agent MD file, how you try uh like everything how you can handle and increase quality of code, it actually gives you a lot of truth. And MCPs as business central MCP is a good example, as you mentioned before, business central MCP, which is a great tool where you would like to manipulate data inside business central uh through your agent, as example, and because of it, you can simplify processes. There's many points, uh testing, testing, you can generate tests uh for A and before, and I'm not I'm not sure if you're aware, I think you're aware, but before uh if a customer asks you to write some or to build some project, and if you will try to try to mention tests and you will say some numbers, they will mostly say no because I don't customers don't see value in tests, direct value. There's value, of course, but they don't see it as direct value. But now with IA you can generate tests very easily, and and you know and not just you can, but I think you have to generate tests. I think test is now uh standard for uh IL. We all need to generate tests, we all need to understand a little bit how it works, but mostly we can delegate this. And uh I like the idea that I listened or heard on uh LinkedIn from Tiny or from uh Vecco that TDD is a great choice for L. Like we can start from tests to be sure tests are correct, and after only after that go to develop something.

You you hit on a number of key points in that whole conversation.

It's yeah, it's uh just in general, but we can focus on something specifically.

We'll get into a lot of those because uh as with many uh individuals that we speak with, I hear some of the things that you're saying, and I just get excited, and my mind goes deeper into it because I spend a lot of time now doing this. I uh I have some questions for you, but I want to make some comments on. We can start with the tail going forward here or backwards. Testing, I agree with you. Customers used to not see the value of tests, but I don't know if it was necessary to tell them that we were doing the test or if they should have just been included in the project. Thankfully, now with AI and the rate of development that we can make, those tests can now be a part of it. Uh, it's interesting about tests. I did some experiments the other day with test TDD, I've been a fan of for a long time because I found it also helps architect the code better because now it forced you to modulize the code to have small tests in the process. And by doing it now, AI can help you do the tests and also plan the development for it from the test. But an interesting thing, uh uh just had a conversation with someone this morning about creating tests, and now the same thing. Everybody's starting to say tests are required, and not only having tests, but proper tests, right? The tests need to fail properly, the tests need to pass properly, not just pass fail. You need to make sure it's really feeling failing for the right reason. But I went back to an old project that had few tests because sometimes it's not always easy to come up with all the scenarios that you want to test for, to be honest with you, because users will do things that you may not think of. So I asked a you know, the AI model or mentioned to the AI model or the framework, excuse me, are there any other tests that I should test for? I'm simplifying it. And it went through and it created or suggested, I was planning at this point, so many tests that I didn't even think of, but were valid tests, meaning the code should pass these tests. I then had it again, I had existing tests.

So this isn't just is this against an old project that you did this against? Okay.

Against an old project that had some tests in there already. So it had my I have an instruction file that we're using now for uh modeling how we want the tests to be created. But I said, go back and can you suggest some tests? It created a whole List of tests, and then the question is here's the plan, this is what I'm going to do, and we all see it. Now that I have the plan, do you want me to go ahead and do this? Right? I said yes. It went through and created all of those tests, and it made one mistake for the set of probably 10 new tests that it made.

It's a suppressive, right?

Followed the structure because I do this in I follow the Garrickin scenario, the you know, the g the the scenario, the give and the win, the then. And it created, I had an existing library code unit with certain functions, and then I had tests broken down by area. It did it.

And I was is this a project that's already live that you went back and ran the test?

Yes. It's an extension that's creative. So we had some tests and it went through and it created the tests for the code. And it just goes to show, as you had mentioned, there is no reason now to not have the tests. And now also um, what I found with testings from a previous scenario is code was introduced to a project that if we had not had tests for it, we would not have identified a problem that it created. Right? So now you have extensions, you have code that works. If you don't have tests for it, if somebody wants to add functionality to it, you may inadvertently, because you're not aware of the flow, and this is where I think you would get into it becomes more important because now AI is just creating so much code that it's very easy to break something else that's auxiliary to uh your project.

So I think it goes back to the value component because I see that you went back to test an old project that's already in production. A end user or client would not see that as like a direct value. Obviously, it's a value to you as a developer because now you're like saving this client from future headaches if someone was to extend it. But I think that's the missing components. Like uh back in the day, whether you test it or not, you create test scenarios, you you uh the clients imply that you're already doing that. Like you've it's like you should be doing that, right? But there's different types of testing.

And I think it's it's too much to when I I think I think testing's important's always been an important, I think it's even more important now. Not again, I'm I'm not trivializing how important it was before. I I've had years of sessions at conferences on creating tests. But I think telling some, I I think saying I'm developing an app and part of this is creating tests, part of it's creating the documentation, part of it's creating this. To a customer, it's hard to assess the value because they don't know the value. They the tests are going to save them in the future from mistakes. But to your point, Chris, I would assume that if you're going to deliver something to me, it should work. Why why why should I have to say test? So it's more of here's what we're doing, this is what the cost will be, is it of value to you? And I'm going to give you documentation which AI can help create. The testing shouldn't even be it's almost like saying I'm coming over to your house, Chris, but I'm going to walk over there. Like, does it is it really even matter? It just matters that I'm going to get there.

From my experience, customers some sometimes try to uh have more detailed numbers, like they try to understand uh why you have numbers, why you have this amount of foul spending. And and if you will say I spent some amount of files to test in the past without IA, they will be not so happy, you know.

No, I understand. I I've lived it and I understand it, but it's almost a matter of somebody needs to test it somewhere, whether it's automated testing or developer testing or user testing. No one should implement something without testing.

Yeah, yes. My favorite type of testing is production testing, right?

There are many implementations where that was their model. They they assumed that the developer uh developed the extension properly and they just threw it in, and then they get mad when nobody tested it. And it's it's uh interesting. So I mean, so the testing is extremely helpful, and I think now it's even more important, but there is an offset because of the speed at which we can create code, we now can include more such as testing, extension documentation, and those infamous guides that nobody used to write.

Yeah, and I think together with it, it actually means you are not losing quality or almost not losing quality, because on top of uh fast generation manual line of code, you have tests, you have documentations which you don't have before in the past with without artificial intelligence. So actually, I think uh quality may be even better a little bit if you have processes in correct way configured, you have CI CD, you follow all best practice with IA, with working with IA, it actually can even uh increase quality.

When AI Adds Work And Risk

Yes, absolutely. So another thing you went back to talking about was the amount of code that we can create um with AI. And testing, just to come back and loop up a testing, testing is not just extensions that partners may deliver or contractors may deliver. If you're a developer for a customer, some some more implementations have their own internal team that can create extensions, they should also test as well. The testing is not specific to deliverables from partners or outside agencies. Even if you're working internally, you should take the time to create those tests for yourselves and you get better quality. And as I had mentioned before, you can make sure that any changes or additions that you make won't impact your existing workflow or your existing code. Uh even to the point where if you have multiple extensions, I always say run your test for all of the extensions as well. Because I have found, and this is uh where we jump into AI assisted development, where have you found AI to be helpful or not so helpful with development? Have there been any situations where you can say that AI for sure saved you a lot of time? And other cases where you said AI probably created more work than it saved?

Good question. Actually, I have several examples where I have many examples where it's helped me to save my time. For example, when you write tests, as you mentioned, it's actually a good uh point to show where tests are save you time time dramatically. Um in my experience, but it's more interesting to to to speak about where it's actually uh create some problems for you or it can take your time. Uh and first of all, uh it can create you some uh difficulties when you uh collaborate with other people who use IA, but they don't uh understand properly how to properly use it. They are not using um uh specific uh uh best practices with agents, they're not uh understanding how it works, they think it is like a magic and uh they don't provide enough context to agents, and uh the quality of this work can be uh less than you expect. And as open source maintainer, uh I realize that sometimes people come to uh repository, to my repository of my open source project, and they try to uh create pull requests, uh contributions to uh to my uh this like for example that editor tool, and they generate it via agent without proper specified context of problem, without proper specifying uh what is a problem about, what is it we you try to solve, and agents generate something which can be like matched. And I heard that many uh open source uh maintainers encounter issues that they are overwhelmed with number of pull requests, with number of contributions which people try to bring to uh projects uh because of IA, because I just generated, and they are not understand how AI works, and it creates some overload without um necessary uh purpose. It's one example where it can be a problem.

Planning Mode And Context Discipline

Yes, yes. Uh I I've heard that as well with open source projects that a lot of AI pull requests are being made, and it's difficult for the maintainers to keep up with those changes to review them properly. It's interest, it's it's it's an interesting time that we live in. The stepping back into AL development with AI and some challenges that you may have had, or any suggestions for someone that may be working with AI and AL development uh when they're creating extensions. What are some things that you have found, whether it's working with ISVs and how AI could uh help with that, or any other places where you notice some challenges using AI?

Yeah. I see some I got some patterns from my experience when I work with AI, how I use it, how I try to get most possible value from it. Uh and first of all, I think you always have to start with planning mode. Like if you um have some problem or you would like to build something new, uh like totally entirely new application, do not start quote or ask uh uh agent to quote for you. Focus on plan. Ask agent to plan. Like give all possible context which you can give, like what you can know about your problem or about idea of what you would like to implement. Give proper context to model. Try to uh highlight what specific if it's some modification of some exist application in uh ISV of a PGA extension, highlight specific objects which you think are um related or can be related to this problem, highlight why you think about it. Try to communicate with agent as you are communicate with uh person, like try to understand that it can magically understand uh get all information if you are not specified. It can, but it will lose context window, and quality of uh generation will be lower and lower because context management is very important when you work with agents. So you try to provide as less but as much as possible uh context because it's actually like you're always fighting with uh two different uh problems. Like one problem you provide not enough context, so agent cannot generate something for you. Another one you provide too much uh context and it's like overwhelmed and it's unfounded on uh your specific idea and task.

So that's a key point I've I've heard from people. It's less is more. Don't get into please thank you and keep adding additional words or even try to tell it what to do, in a sense. If you the more boundaries that you give it, it seems to have more of a challenge with the output.

I agree. And when you set some boundaries, sometimes if you're very sure about it, it can be useful. But many times I realize when you set some boundaries, like you try to lead agent in some specific way, but you can be sure that your way is the best way to do things. Like you have to learn from it as well, and try to try to look what it suggests, and then you will see all possible uh branches of solutions how you can solve the problem. Uh you can like choose the best one. Um so plan mode, context management, uh session management. Of course, you don't say I I I I saw this mistake very often in the past. People uh speak with agents in one session, they're not pay attention on context window, and uh agent just overwhelmed with all different specific tasks which uh people try to manage in one session and screen problem. Start new sessions, try to manage it. Um what else? Also, it's important to choose right tool because there's many benchmarks in the internet which they specify, like we are best agent or we are best model for specific agent. Uh it's hard to uh predict what is better for AL development on those benches because they are mostly based on C sharp, Python, on other languages. But uh recently Microsoft released a BC uh bench uh for agents and they uh it was as I remember they only count two agents, quote quote, and maybe uh codex, not sure about codex and it's one point when you can look on it. Try to find IL benches to understand it, but actually what I will uh give one small advice to everyone who is interested in it is try to yours try to use it by yourself, try to use different tools. Uh because if if it's something good for one person, it can be not good for you. You have to choose, you have to try, and you have to spend time on it, and and it will give a lot of value uh later when you will realize what tool is better. Because there's like cursor, uh codex, cloud code, uh uh copilot, all different tools what you have. But my honest answer, try it, and you will see the difference. Try to compare it by yourself, by your internal tickets. I even have some small internal um tasks, and I try to uh evaluate those tasks every three months or something like that. Six three months uh on all uh popular agents or clients which I can use, and I can see the situation is changing. Like six months ago, codex was better, six months ago, compilot was better, and so on and so on. It's constantly changing situation. I prefer Clot Code, I even uh wrote about it uh several times. Um but I'm not sure if it will be forever. Like maybe tomorrow it will be better, something from Google, who knows?

Like so, one of the things as you're going through your development workflow with AL development is not all models or frameworks work the same, and what may be good for one person for one task may yield different results for somebody with a different task.

Yeah, and also it uh depends on how you communicate with model. Every parse person has a unique style of communication, right? And it also affects uh how uh how what output you will get from agent from models.

It's a challenge to keep up with.

For the both of you, so how does that work when you have uh when you're working on a project together and you have maybe three of you working on a solution, and then they may have someone that's a third party, outsource whatever, uh that a client hires. Do you all use different uh framework? And how does how do you maintain that? Like how does that work? What's the workflow for that?

Uh from my perspective, I have such experience um very often that people use different tools and agents. They uh one developer uses codex, one developer uses copilot, uh, one developer use uh copilot inside Visual Studio Code, one is using copilot uh as uh command line, right? Technically, it's not so important that people use different tools because the main things to get the best output from it. Uh I think like even if you are using two uh different tools, but it gives you good output, it doesn't matter what tool created. But how you uh communicate with different agents uh and your experience, it can affect quality of work what is uh will be as output from agent. But at the end, it doesn't matter what was by what tool it was generated. That's what I see.

Yeah, I agree with you. I I uh it's a it's a good question, Chris, because it's not to confuse the tool that somebody's using to create something, it's what they create is the important thing. As Old Me had mentioned, it's the output of the quality. What becomes important in those situations is not the focus of which AI tool they used or which edit IDE they used, it's what they did and how you manage it. How you manage your development workflow with source control management, with branching, with merging, with code review, those are the things that become important. So you can have an internal team that has multiple developers, and then you may need to have another person also contribute, they still can fall into your workflow. What they use to create it's not important. So even if you are a customer of or a user of Business Central, excuse me, and you're managing your own source code and your own uh repository and workflow, you can bring in others and have them fall into your workflow. You just need to have the discipline of how to manage the different branches.

Yeah, I think I think that's a good point because I think down the road there's going to be, I feel as though there's going to be more of that as a lot of clients uh understand that you can have multiple partners working on a solution and as people vibe code, you know, someone has to manage that. And then at some point, I mean, you're right, it doesn't matter what tool they use as long as the output is the the the right expectation or the correct expectation of the output. Um but at the same time you may have some balance of like if you're using a specific framework, then your testing scenarios may produce different things or different testing scenarios versus someone that just you know uh committed their their code into. So it's it's an interesting balance of what that looks like in the near future, uh, especially when your clients may want to be more involved in that.

Uh because you know, actually, yeah, it's uh great points. And uh CICD is what can help us to manage all this uh mass workflow. Uh we just can um try to automate as much as many process as we can. Automate uh test running, automate uh linters, uh automate uh compilation of projects, automate everything of it, and it creates and it starts to be even more it was important before, before Artisan Terence uh was starts to be more popular, but uh it's important now even more, same as testing all the CICDE processes, and uh that's why we all need to look on LGO, how it works, how we can implement it in our projects, and it's actually for free. You can implement it for your uh open source project and for PG extension. Go ahead, use it. And I actually very um I almost all my projects that I work with customers. I always try to say we need to implement CICD, we need to implement LGO in uh this specific project, and I explained why, like uh like what value it brings to customers, and mostly most of the time they are actually supported because they see it's actually very easy to show them uh value of workflows of how automations help. Uh for example, from to save uh project from regression bugs or problems, a lot of things can be um automated, and but I think it's main tool also what we can focus as developers of L developers and to using ALGO, to using uh CI CD processes.

MCP Basics And Tool Choices

Yeah you brought up a good point. ALGO is It is a free tool that runs in GitHub. There could be a cost to the runners that you use, depending on how you set that up for your projects, just to put that in there. There is a tool for it. But I I I literally had a conversation with someone this morning about that in implementing these on projects, that they were having problems with developers building the workflow is to have a branch, review the branch, bring it into the merge it into the main branch, as we'll call it, and then that gets deployed. Whereas if you have a developer that didn't update their main branch with other code changes from other branches, and they just did it with their branch merged in and then uh built the application and deployed it, you in essence can remove changes that were made. So by using tools such as ALGO and GitHub, the extreme benefit is you have the security of knowing whatever was built, the artifact that gets built as part of your CI CD is what should be deployed. You don't have to auto-deploy it, but that's where whomever is going to load it. And I'll be the first to say is I don't think developers should be the ones responsible for loading the changes that they make. You should have someone who knows how to pull down the artifact, and then that's what gets loaded. But that just verifies that the tests were run because the AL Go will run the tests that are within the extension. And it also validates that anything that's in the main branch is what gets built, used, and deployed, and you're not relying on somebody refreshing their branches or refreshing the changes of the code that they have because using Git you have a local copy of it. Yes, it does sync, you know, you do pull and push from the central location for it, but you still have to go through that exercise of doing that. Yes, it can be automated, yes, you can do it, but some people still like to do it manually. So the point that you had made is using tools like that in this world of AI development to me and team development becomes even more important. Not that it was less important yesterday, it's more important today. So you these are some good things. I know we're talking about AI and AL development and open source, but um so it with your AL development and AI, what tools do you find useful to help you with it? Are there any other tools that you have found can assist you in this journey, or even uh if somebody wants to start using it? You had mentioned a little bit ago, just start using the tools, right? Just start using AI for your projects. What are some things that you found were helpful to get you to where you are with your development journey?

Um, yeah, there's several tools which I uh think is important uh to um to start with or to learn or to look on it uh a little bit. Uh first of all, uh I think it's a good point to get some basics. Uh because if you look on uh different tools, as example, like uh agents in common line, uh they all have some different same uh similar uh technologies behind like MCP. What is MCP like if you try to understand what is MCP it can help you to uh get more efficient with any agent? Like you are not focused on specific tool. Uh so I will start. I think it's better to start with some basics, like to understand this MCP. Like, okay, MCP is some protocol which will uh allow my uh IA to uh communicate and to change something on external services. It can be business central, it can be your browser MCP, it can run uh for you some uh business central um page and even try to click some uh buttons directly in uh your um browser if you need it. And all these are done by via MCPs. So start from basics to understand the context windows, MCPs, but speaking about specific tools, yeah, I have several in my mind. So I started uh originally with uh um cursor, if you heard about it. Uh it's like uh form of Visual Studio Code, but with implemented um models of AI. But it was like in the beginning of the uh my journey, and um I realized that now you are not needed uh cursor because everything works inside Visual Studio Code. Uh naturally you can have uh several different models. So now I think it's good to start from Visual Studio Code. It's our daily uh EDE where we as L developers work every day. So it may be uh good starting point to use uh copilot inside Visual Studio Code to select some models. You can select different models. You are um you can select uh anthropic models, you can try open eye models. Try to mix it, try to to start with EDE. I think ED is a good starting point. Uh next, what I uh think is worth to try after you focus on some Visual Studio Code. Um I think it's next this will be useful is uh common light agents. They are independent from ED, you can run them inside your terminal. And there is some value behind it. First of all, you can run it everything there. Next, they are working this file system differently than ED works. It works uh per my experience more efficiently with um all these uh commands which you can execute inside your uh terminal, like bash command, uh PowerShell command. It's very easy to automate, and you as a developer can easily handle it. Um so try some command-line agents, and there are several popular agents on the market or in general. There's like uh compiler agent as well, there's uh codex from OpenAA, and there's uh Cloud Code uh from Anthropic. Uh I try to use them and try to mix a little bit to understand what tool is better now to what is better than now because it's changing dramatically, as I mentioned. So try with Visual Studio code ED agent inside it, try to um read the basics. What is MCP, what is context window, what is uh slash comments, or uh uh what is um MD file skills, it's it's actually called skills now. Before it was slash comments, now it's skills. Uh how skills can affect your um generation, agent generation. Try to next do next steps to use some MCPs actually to on practice. Next steps try to use comment line agents, try to compare different agents to see what uh value uh you can get from it. I think uh also try business central MCP, it's also a good point to check uh how it works, how data is flowing, how you can just tell to agent, please show me or delete customer numbers. Of course, it's not for production, you just do your daily work on some sandboxes or on some uh uh Docker environment. So I think those like main entrance points that you can go.

It does sound like there is a lot, but as with anything else, it sounds worse than it is. And I think once you start working with it, it all comes into play. And I think the important part is to start using it and not try to know everything, just know what you need to know to get kind of work your way through the task and see what you need to do for that task to to be uh uh be successful with it.

Yeah, and I I can tell you practice and practice. Like if you just will read theory, it's will not be so useful. Try to do it in practice. I realize by myself that when I do practice, I um start to understand things better. But when you just read theory, yeah, you know MCP, but try to it in practice, and you will realize uh actually how it can be useful for you and how it can uh give you some output, valuable output.

Why Business Central Knowledge Matters

It's it's um yeah, it's a strange world now with all of this output and uh business central developments coming along, and the business central tools that they keep adding are also nice to have, and it's nice to see that they're adding all these tools to assist with the development point of view. And in my opinion, it's also reinforcing that the days of just development are gone. I think that developers need to have a good understanding of the application and the function of the application so that they can work with AI to develop these solutions. And also, as you had mentioned at the beginning, the importance of still understanding and the ability to know the architecture and the code. So it's not it's not one of those things that, oh, AI, I can create this extension, it's going to do this thing. I think you do need to have a sense of understanding and review how to review what gets created. Uh, I think I keep saying business central to me is a little bit different because it's in a box. It's a predefined application that you need to work with within the constraints of how the process and the application works. We can't just change a complete application, every build is a standalone application, it has to work within a framework. So it's important to be able to understand that framework so you can review what gets generated and then deliver something of quality.

It's is that I actually agree, and you bring a great point. Like as business-centered developers, we are not just developers like um Python developers who just know language and know all these um keywords and how to do and write some algorithms, but with business centered as business-centered developers, we are also should and we actually should really and must know the basic functionality of business center, how it actually works right now, how it actually works from a functionality perspective, not just from code how it's coded. It's one part, and but understanding how it works for different types of businesses, how you can configure it, it's actually super important in uh any era, even before I it's also was super important. Understanding of how it works, it's actually because if you without this understanding, you can build efficient tools, you can build efficient uh uh solutions for your customers because you will not aware how to handle it in the best way. That uh you don't need to quote anything what you hear what the customer asks you. You need to think, okay, is this already a functionality existing business central? Can you just implement just use it without coding anything else? Because I saw in my experience, many poor developers try to build something immediately when someone asks them, like, please create me this and this. No, try to think if this is actually required to build, if it's actually maybe already a part of standard business central library, maybe you're not aware of it. Try to do some research before you say something. Try to think. And yeah, so I think understanding of functionality is important, and now with AI, it's even more important because now we have more time for architecture and also more time for understanding functionality of business central as L developers. Like now you cannot say, Oh, I'm just code. Noah, it's not on, man.

You hit that well, and you answered the question, and I think it's a great way, I hate to say it, it's a great way to wrap it up. What do you do with all that free time? That's the question that everybody's saying is okay, with all this free time, I don't have free time. Don't look at it as you have free time to go uh you know, not work as much. Now you have free time to, as you had mentioned, learn the application. You have more time to understand the architect, the application, more time to architect and plan a solution. So it's not necessarily that we have all this free time. I can get work done in two minutes and now I'm done for the day. It's I can deliver better quality work. And this is where that question gets asked. What do you do with the time? You take the extra time to add that extra quality to it. And again, not to say that quality wasn't important before. I'm not trying to discount it, but now you can deliver a better quality product in a shorter period of time. That to me is the answer to all of this. It's better quality in a shorter period of time.

I fully agree.

Where To Reach Volodymyr

That's a good one. That was a good, good with that.

I think uh would like to thank you for taking the time to speak with us today. I could talk about this stuff, and I'll probably now I'll just have to put you on my list of calls to have that I could talk about this stuff all day, every day.

Because we can spend days and hours to speak about it.

It's actually board topics like oh yes, and it's uh you're in some of the same chat groups we're in, and we get to talk about this a lot, and uh, you know, it's the wonderful chat groups because I'm in the United States, most of you are in Europe. I can wake up in the morning and there's like 65, 70 messages I have to read through and screen through. And it's just like you know, really what's going on. But it's it's it's a it's an exciting time for everybody. And again, thank you very much for taking the time to speak with us today and sharing your valuable insights. I think anybody who's in the AL journey, uh, we're all in the beginning of this journey. Anybody who says they're an expert on it, I mean, how can you be an expert for technologies that's so new and it's changing? So anybody that's uh currently using AI, hopefully that uh thank you for sharing so many insights. Somebody that's new to it, you also shared some insights for that as well. And if anybody would like to talk to you more to learn about your open source projects or maybe any other AI for AL development tips that you can share with them, what's the best way to contact you?

Sure, um there are several ways. Um you can have just uh comment and put a comment in uh my blog. I have a blog uh where I uh try to share my um knowledge, what I learned from uh my experience about development specifically for business center development, also some some IA insights and tools. Uh you can comment it. You can contact on LinkedIn if you want to ask anything. I'm very open-minded. If you just just ask me on LinkedIn, I will immediately answer you, or when I will have time, of course. Um, so LinkedIn, my blog, I have also um uh Blue Sky. Anything of it will be okay to contact me.

Uh great, thank you. And we'll have your contact information with the show as well. Uh again, thank you for taking the time to speak with us. Look forward to speaking with you again soon. Uh enjoy the rest of your day and have a great weekend. Talk to you soon. Chao ciao.

Take care.

Bye.

Thank you, Chris, for your time for another episode of In the Dynamics Corner Chair, and thank you to our guests for participating.

Thank you, Brad, for your time. It is a wonderful episode of Dynamics Corner Chair. I would also like to thank our guests for joining us. Thank you for all of our listeners tuning in as well. You can find Brad at developerlife.com. That is D V L P R L I F E dot com. And you can interact with them via Twitter, D V L P R L I F E. You can also find me at mattalino.io, m-a-t-a-l-in-o.io, and my Twitter handle is mattalino16. And see you can see those links down below in the show notes. Again, thank you everyone. Thank you, and take care.