Building Better Developer Habits - Embracing AI for Efficiency and Growth

Welcome to Building Better Developers, the Developer Noir Podcast, where we work on getting better step by step professionally and personally. Let's get started. Well, hello and welcome back. We have already gone off the rails, but we are back. Building Better Habits, Building Better Developers. This is the Developer Noir Podcast. In this episode, we're going to talk about AI, and this is not just to be cleared, an AI generated podcast. Now I have to introduce myself. I am Rob Brodhead. I'm one of the founders of Develop Noir, Building Better Developers. I'm also founder of RB Consulting, where we are a boutique consulting firm. We come in, help companies find the best way to use technology to build. Maybe that includes like maybe the people side of it. Maybe help you build a technology team, a software development team. Make sure that your processes, procedures, software, all that good stuff is there through simplification, automation, integration. We find ways through all of those shuns to help you reduce technology sprawl and make the best use out of your technology. Now in this season of habits, we've had challenges along the way. And of those challenges, I've talked quite a bit of particularly the to-do lists and the Pomodoro pieces. I want to say automation is one that it has really been helpful to keep that sort of like more forefront in my mind. It is very helpful to have, as we've said early on, we had the seven days, like look at what you're doing and in the seven days do some sort of automation. This has been something that I've continued to do and it does keep me on a regular basis thinking about like, what am I doing? How am I doing? Can I do this better? More importantly, can I automate this? Or in some cases, can I remove it entirely because it's really a waste of time because I could automate it and just it gets done or things like that. So that's one that I've been really happy with. That's a challenge that I think has really helped me move the ball forward as well is to keep in mind on a regular basis, where can I automate my life, my work, the things, the tasks that I'm doing on a regular basis? Good thing and bad thing. Good thing is last week or two, I have been like full busy. That is always good. I like my job. I like to be like getting up in the morning when I'm no like, I got to go. I got to get stuff done. I got crap. I got to do today. The bad side of that is that is totally blown up like my sleep, sleep schedule and stuff like that. Now my internal clock always has me getting up at the same time. Doesn't matter what day of the week it is. Unfortunately, it also doesn't matter when I went to bed. So if I go to bed at eight o'clock PM or two AM or five AM, I wake up at five AM basically. So I better make sure I have enough bandwidth to get some sleep in before my body says, hey, it is time to get up. Now I can groggily put myself back to sleep for a little bit, but guess what? Minutes later, my body's like, hey, you fell asleep. You need to get up. It's not a lot of fun, but what is fun is getting the introduction from Mike every time. So go ahead and give us your non-generated by AI introduction. Good to see you, everyone. Yes, we are not robots. My name is Michael Molloch. I'm one of the co-founders of DeveloperNUR. I'm also the founder of Envision QA, where as the title of the company implies, we offer lots of quality assurance testing services to help companies that either don't have a QA department or are having trouble with buggy software or software just doesn't quite work the way it needs to. And they need some help getting back on track. That's where we come in. We help you improve your quality assurance of your software, be it through manual automation or integration testing. Goals. A lot of the things I've been focusing on recently have been kind of working on my planning and scheduling to kind of get back on track a little bit with kind of where I'm going with things, especially with our last discussion on Agile. I've been trying to work through trying to make sure that I'm following the ceremonies correctly, doing the planning properly and really trying to prepare for those meetings. So it's kind of a combination of planning and scheduling and meeting prep, making sure that I'm showing up on time for the meetings. I'm focused and I'm present during the meetings. Good and bad. Good. I've been focused on our goals and our challenges to build better habits. Bad side of things like you, I've been getting a little bit busier, so I have to be a little more thoughtful as to what I'm working on right now to stay on track. So I've been kind of going off the rails a little bit in a couple of different directions, losing sleep, like you said, but overall more good than bad this week. So let's talk about AI and do it in a non AI generated way, which is sadly enough, how a lot of people do this these days is you'll have stuff and they'll say like, here's how you use AI and they're basically just like throwing some stuff out and go, look here, I could, AI can do my job. Now AI as we, we actually threw some stuff in the also just throw out, go check out the YouTube channel. We do have a lot of information out there and you will understand why sometimes we come into this and we're a little bit like punchy before we've even started because sometimes the pre-show gets a little bit like out there or maybe we're just out there all the time. But AI, I think the problem with AI is or with using AI and I think this is where we're going to, where we're going to go with the challenge and where we're going to build a habit is using it as a tool that can help you do better what you are already doing. I don't think AI, well, yes, if you're not a coder, AI can generate code. If you're not an artist, it can generate art. We'll call it that because everybody's seen, yes, there is some, there are some AI art generators that sometimes the art is very good. It's exactly what you need. Sometimes it is horrible. And so you have to know the difference. Now it's the same thing with if you're writing like grammarly and some of these tools that are really good about applying rules like grammar rules, spelling rules and things like that. Those are perfect with AI because it's usually going to give you a pretty good feel for that because if you can, the whole point is if the AI has the right rules, the right constraints around the configuration around what you're trying to generate, it's going to do a pretty good job because what you're really doing is you're just automating a task that you do. And so if you can tell it how to do the task, then it's going to be able then it's an automation for you basically at that time. That's why code is a really nice example because you just, you know, you can put out comments and say, I need to do this, I need to do this, do this, do this, do this in this language and this style, blah, blah, blah. And it'll generate something out. Now, it's probably not necessarily, it's not always going to work. But if you know what you were going to build anyways, now you're just checking somebody else's work. You're making some changes to it and you're able to move forward. The same thing with writing. And this is where like I think the biggest thing I've seen that is useful with this lately, and there are a lot of areas, is actually in summations of sorts. So, for example, we, when we go through, we use Zoom for our stuff, we get done, we get a summation of what is what we said, what we cover. So it's really great, particularly if you're using this for meetings, that you can come back and go, OK, well, what was this meeting about? Without having to look at the recording again, you can get a summation and say, well, this is really what everybody talked about and how they went with it. It may not be perfect, but it will give you an idea. And then it's something like, oh, I need to go back and like dig into that. Or cool, it covered, it got me the couple of points that I wanted it to get. Same way as there are tools out there like Descript and things like that, that will take an audio piece and convert it into using AI, they will convert it into text. And so it does give you a better. The additionals of AI give you a better way to convert because it's doing some of the smarts that in the past you would have had to do. So, for example, it can take a conversation and it's going to know because it's got some smarts to it and things like that. It's going to know, for example, like Rob and Michael are the two talking, and it's going to be able to tell because of the differences in the voices. Just as you do in your head, it's going to say, you know, Rob said this, Michael said that Rob said Michael was wrong. Michael said, I know I'm always wrong. I'm sorry. I made a mistake. Okay. Maybe not like that, but unless it's an evil AI, you got to watch out for that. Cause sometimes the AI might beat down on you a little bit because we all know Terminator is out there somewhere. But the point of this is there is a lot of, there's a lot of promises about AI. Like there have always been just like there have been promises of a lot of new technology over the years, but there is a lot of reality that is there. We have bigger machines, bigger processors, more memory, more data that's available. And particularly when you're finding tasks that would be monumental for you to do, because they're very repetitive or very rote, you probably can get AI to do that for you and get through and automate what you would have done. So I think that's where instead of artificial intelligence, maybe you think about it as like automation intelligence, then you're going to see where I think we're really going to be able to leverage it and use it to do our jobs better as opposed to treat it as a crutch to do our job for us. Now I am not going to use AI as a crutch. I'm going to actually use Michael so he can continue this conversation. Thoughts? Yeah. So that last thing you touched on immediately made me think of Dragon naturally speaking. That software has been around since the nineties and how it typically works really kind of simulates what AI is today. You download the software, you install it, you have to sit there and go through a whole bunch of recordings, well, of word things on the screen. You have to speak it. It records what you say. And essentially you're training the tool to be able to understand your voice and be able to dictate what it is that you're saying. Now, of course, today we've got better tools like that. We've got Descript, Zoom, and all these other tools, but all that came from that concept, that idea, what it is, is you are giving it, you have to train it. You have to give it some rules in order for it to get it right. Now AI has already spent a good deal of time before these apps even became available to the public, going through scraping millions of websites out there, looking at different documentation, looking at different topics, different ideas, categorizing all this. And essentially what it did is it built this very procedural database driven chat system. So you type in a topic and it's going to go search very quickly through all of these libraries based on all these rules and regurgitate back some content that today now really seems like it's reading your mind. It's very quickly giving you back very reliable and very consumable information. Now, how does that really work in our jobs in the real world? Like how can we use this? Well, as developers, because there's so many different sources of information, we've got GitHub, we've got Stack Overflow, there's all these great sites that have code out there. Especially if you are a Microsoft fan and you use Copilot, they've already scraped most of Stack Overflow for code and content. So if you search a topic in Copilot, chances are you're essentially searching Stack Overflow and it's going to give you a very quick summary of probably something someone has already answered. So let's say you're looking at a topic and you're looking at a topic and you answered without you having to go through all the time and steps of reading all the different things on Stack Overflow. I'm a big tester. I like writing test cases and I've actually found that chat GPT, although IBM's working on a great tool and not Red Hat, who's the ones that do PyCharm, IntelliJ. Oh, JetBrains. Yeah. Um, they are, their AI is getting pretty good for code as well. So between them and chat GPT, nowadays, if you just need help building a stuff, remove all the boilerplate crap that you need to put to just get started. If you literally go out to chat GPT or most of the AI tools between IntelliJ and chat GPT type, Hey, build me a project for a webpage or like Python. Boom. Here's the stub, export it. You've got a project you're up and learning. In Java world spring and like Eclipse and IntelliJ, they've been doing that for years. You can now go into most of those projects and say, Hey, I need a project for X. You click a button. It automatically builds you the stub, most of the project. And if you're lucky, you can even say, Hey, include examples. And boom, you actually have a working project. You just need to tweak. Well, AI can take that one step further because these AI tools have scraped most of the web, chances are someone has done something close enough to what it is you're looking for that a typical Google search that gives you multiple sites to look for in any of the AI tools will actually give you some code examples that may work out of the box, or will give you enough of an idea that you can either go Google based on what you see to go dig a little bit deeper or ask AI or train the AI to, Oh, that's not quite right. I'm really looking for this. And granted, I will say a year ago after your third or fourth AI search, the responses got garbage and you started getting really lost in the weeds today, though, if you spend about 10 to 15 minutes, really refining your search through AI and communicate back and forth, it will start understanding what it is. You want, you have to make sure you tell it when it's wrong. Otherwise it's going to keep giving you more of those responses, but it's just like Google search. It's like, I go out and I like search for bathtubs. Oh, I don't want bird baths. So minus bird bath. And you basically just have to understand the tools that you're using and get really good at it. The last little tip I'll throw out is testing. Now my company really focuses on testing and quality assurance. If you do zero testing, or you really have developers that just don't know how to test or don't like writing unit test, make them use an AI tool just to write simple tests, it may not be perfect, but it, something is better than nothing when it comes to software testing. I have to agree completely. And honestly, it's, it's usually going to be in testing and things like that. Those kinds of automated processes are, it's not going to be just something. It's sometimes it's actually a very substantial something and it saves you a substantial amount of time. So it is actually very impressive. I will throw out there another one that I have used that from a code point of view that has gotten really good, really quickly is whisper. It's Amazon's version of this. All of these tools are available. If you use visual studio code, I think all of those are plugins that you can use. Otherwise, if you use whoever's, you know, most IDEs now that have got some sort of plugin or module, you can find it for these tests it out. These are things that sometimes do become a little bit frustrating because they will like just blow out a whole bunch of code when you didn't want it to. And then you run to why you suddenly had 4,000 lines of code generation. You're like, I don't remember typing that. And that's why. So you may have to like undo and some stuff like that. But that is the key to these things is like these work, these are getting better. And if you either teach the tool basically, and give it some, some, some sort of interaction and help it to give you, you know, help it to get a better solution, then you will end up with sometimes a very solid solution. And if not, at least something that puts you further down the line of being ready to give your solution array. Now, before we get into challenges, I do want to mention it was January 4th of this year, funny enough, that, which is 2024, you can go back and you can find out our interview with Chris Barkhurst. And we talked about AI and some very good conversations about like how his company was using AI to do some things that are maybe a little bit out of the ordinary, but also I think very telling about what AI can do for us now. Now, from the challenge point of view and the building a habit, the habit here is this challenge is actually going to be pretty easy. Go to whichever AI you can just, you know, go to a search engine and find an AI tool. You know, if you want to just chat GPT, great. But if you want to use a different one, great, go to their site or hook them up, you know, connect in and just one time a day, take a problem that you're looking at, particularly if it's something that's got you like, when you hit a pause, we're like, you're writing some code. You're like, how do I solve this little problem? Take that problem, write it up, throw it in the, in the AI and see what comes out. It may be junk, but it may be exactly the solution you need. And it may be just saved you an hour or two, something like that. Now, I don't think it's going to help you with configuration issue problems, which is the ones that eat up more time for me than anything else, but you never know. It may give you exactly the thing that you need, particularly if you say, why doesn't this work? And you give it to it. Sometimes AI will literally point you exactly to, Hey, you forgot to do this. Or it would give you a list of things to say, did you check this? Did you verify that? Did you do this thing? Just a lot of ways to break you out of your mindset and your mold, because AI is going to give you something a little different. And a lot of times just that thinking outside of the box will help you move on to the next step. So use it as something to like, you know, as a, as a way to help you along the way, as opposed to a crutch where you're like, Oh, I'm just going to throw it over the wall and see what AI gives me instead. Work your way through it enough or say, well, I think this is how I'm going to solve it. Give it to AI. You're going to be a, you're going to get a bunch, much better response and probably something that's much more useful based on that. Sort of like if you shoot us an email at infodevelopmentorder.com, you will get more useful episodes in the future, at least to you, because you will give us feedback and we will make sure that that is part of what we take into account as we do future episodes. And right now we're bumping up against the year end specials for 2024. We will come in and we will wrap up this season in 2025. And then there'll be a new season, which you will probably be shocked, but if you followed us for any period of time, you will know that we haven't even figured out what that's going to be yet. So you can influence future seasons of Develop-a-Nor if you send us an email or leave us a comment, whether you're watching this on YouTube out at the Develop-a-Nor channel there, whether you, wherever it is that you get your podcast, you can leave us something there and go out and we've got a form on thedevelopmentorder.com site. You can hit us up. We've got a Facebook page. We've got a at Develop-a-Nor on Twitter. We're out there and luckily there's not a lot of Develop-a-Nors, so you probably will find us. If you are on Alexa, then you just, you have to call us Building Better Developers because she doesn't understand Develop-a-Nor because her AI was not there when we got started with all of this stuff, but I digress. That being said, I want to wrap this one up. So go out there and have yourself a great day, a great week, and we will talk to you next time. Thank you for listening to Building Better Developers, the Develop-a-Nor podcast. You can subscribe on Apple Podcasts, Stitcher, Amazon, anywhere that you can find podcasts, we are there. And remember, just a little bit of effort every day ends up adding into great momentum and great success.