Keeping Forward Momentum When You’re Overloaded: Small Wins + AI Guardrails

Welcome to Building Better Developers, the Developer podcast, where we work on getting better step by step professionally and personally. Let's get started. Hello and welcome back. We are continuing our season where we are getting unstuck. We're moving forward, getting forward momentum. We are getting out of the starting blocks and starting our race. And sometimes you start at a little bit slower pace, but we're going to try to do our best to get a really good start at any rate. What are we talking about? We're talking about building better developers, the developer podcast. And I am talking about myself for a second, Rob Broadhead, one of the founders of developer or building better developers, also the founder of RV Consulting, where we help you with a technology reality check. Take a look through your junk drawer, clean that stuff up, get you a roadmap for success and moving forward. Good thing and bad thing. Good thing is I have made it through January at this point. It has been a crazy month, like weather issues, moving issues, all kinds of stuff. It's just been a really crazy month. So the good thing is I have survived the month. On a bad note, though, doing all this stuff and being very busy, I have really gotten off of my physical fitness kind of routine and stuff like that. I'm running around doing a lot of stuff, so I don't feel like horrible. But also, I think I could do a little bit better. So it's one of the things that it wasn't on my roadmap. My radar for the beginning of the year is maybe a resolution was to just make sure getting my regular routine in. But now, as I get into February, maybe I need to do a February resolution and do that. Another thing I resolved to do is stop running my mouth and go ahead and pass this over to Michael so he can introduce himself. Hey, everyone. My name is Michael Molloch, one of the co-founders of DeveloperNUR, Building Better Developers. I'm also the founder of Invigin QA, where we build and test custom software that eliminates the bottlenecks. That way, your business can run smoother and grow faster. Good thing, bad thing. Good thing, the snow is finally going away. It will be gone hopefully this week, and life will get back to normal. Bad thing, similar to Rob, because of the snow being snowed in for 10 days plus days ice in, I should say, not snow. You really don't get to move around too much. Yes, you can move around your house, but that is very limited. However, you tend to be even more limited if you try to go walk on ice to chase your dog and you find out that living on a hill is not fun. Once you slide, you are stuck till you figure out how to crunch your way back through solid ice. Yes, just yet another way that you need to figure out how to get unstuck. A nice little way to segue back into our conversation with Andrew Stevens is going to continue. This is part two. We're going to pick up right where we left off. This is get like, get your pencil and paper out because there's plenty of notes to be taken. He drops a lot of great knowledge bombs. I'm looking forward to you guys also experiencing part two. Here we go. One of the things we haven't discussed yet, which I'm curious, is given a lot of the similarities to your style that I see with mine, how do you prevent yourself from burnout? How do you prevent yourself from reaching that cognitive load where you have too much? It's like, yes, you're still getting things done, but you're slowing down. Things aren't getting done as completely or you're just not hitting your goals. You're not hitting your deadlines. You've either taken on too much or things just aren't working. How do you reset? How do you work through those kinds of times? Yeah, I think every problem has its own focus or own method to resolve. Again, I'll talk about that. I'm lucky to have a partner that I can work with. Sometimes her interests align with mine and we will carve up a particular issue or problem and work through that. So I'm lucky to be personable. Sometimes I'm working with great friends and co-founding things with me so I can work with them from that level. I think it's about communication initially. It's okay to be busy and it's okay to be behind, but as long as that's communicated. That's one thing, again, that it's very easy to get lost in your own thought and lost in and forget that people are not necessarily aware of the issue that you're working through. So you need to communicate. You need to be upfront and working with your stakeholders of any kind. So that's always important. But when I'm working on my own product for myself, how do I handle that? For me, I'll try and pick something small and make it achievable and work through that. Sometimes you look at a big problem or a lot of deadlines, whatever, and you go, how am I ever going to do any of this? So I'll try and achieve something small and do that. And that makes me feel better about the issue or the item. And I'll do that on whatever is in my queue or whatever I've identified. Sometimes there are things I just don't want to do, business administration. You just put it off to the last minute and I hate it. We've got tax deadlines coming up soon. And last year was a busy year, so I delayed my tax returns until October. And now it's something on top of me again. And I'll never do that again, because usually you've got a whole year to wait. This time it's only like six months. So I'm not going to do that ever again. So some lessons learned, I've learned to delay life. But you need to push through the compliance things. And those are always the hardest I find. For me, it's fun. I can always find more time to read that new technical thing that's out. Like Mechtasik I mentioned earlier is something that I'm really into at the moment. I'm looking at these alternative networking methods and things like that. So there's always plenty of time to look at the product and look at what is on Kickstarter or whatever in the product space. And that's always a lot of fun. So for me, a lot of engineering things are actually downtime and it just continues to build. So I'm looking from that perspective. How do I handle my time? Sometimes I do get close to burnout. I'll just keep working those hours and I'll get to a point where I just can't sleep anymore or I need to do it. And that's when it's good to have people around you that will tap you on the shoulder and just say you got to back off or whatever. It's good. It's something I've got to guard against. So I don't have any solution for everybody. This is something that I'll try and manage by breaking big things into smaller chunks to feel more successful from a mental perspective. Motivation wise, those small things, those small achievable things help keep me motivated. The big demotivator is compliance and taxes, getting those done on time. I think we all hate doing that paperwork, but it's got to be done. So I think those are the two opposite ends. The cool tech stuff and the terrible compliance stuff are the opposite ends. You've got to have the right stuff. You must do. That's why there are very large accounting companies that make a lot of money because you've got no choice. Then there's all these poor struggling tech people that it's optional spend. So that's the balance there, I guess. Rob likes to talk about eating the frog all the time, get the hard thing out of the way first. And sometimes that's possible. Sometimes you got to break it down into smaller pieces to kind of work your way through it. Interestingly enough though, through your explanation that you didn't touch on AI, so are you not using AI at all to kind of help you organize things, plan things, or are you just kind of keeping that out of AI and just using AI strictly for engineering and software development? Yeah. So I use engineering, like I said, for software engineering for sure. I do use AI. I'm using some great tool in Gemini Enterprise that helps me with my calendar with a client or two. So it'll look at my calendar and just say, I'll get it to give me a daily rundown of what I need to do today from the client's perspective. It'll give me a summary of emails overnight, my instant messages, my JIRA tickets, all those kinds of things. It'll give me a basic summary of the day. I always take that with a bit of grain of salt because LLMs get it wrong. But it will give me a decent summary and I'll use that as a basis for my customer that has that enabled. I'll do that, absolutely. But I use nothing across the customer. So I try and keep things siloed and isolated because there's still a bit of a trust factor with AI. I definitely don't want the wrong data into the wrong AI model in history, some way that is not appropriate. I do try and keep that separate and I will in those allocated times, I'll use AI specifically for the customers that have it. So yes, to a degree, I do do that. Yeah, I would say that's like that's probably one of the best. I've said I was actually having a discussion every day is I think my AI spend has been one of the best spends I've had on a tool in probably ever. And part of it is very valuable. It's very helpful to have that ability to just like have folders and have stuff completely separated out and be able to sort of not have those things bleed over. However, I have a few checks along the way. So if something does bleed over, then I'm like, hey, wait a minute, like you're getting out of your way and you're touching something you shouldn't. And it gives me some heads up if I need to go dig a little deeper. But I do want to swing back because I think there's two things. There's like one, I think I've talked a lot not only about like eating the frog and just get the stuff out of the way that you don't want to do, but also the value of incremental improvement of just like get a little bit each day. And I think it is I think sort of to follow up your thought there is the. That little bit, even if you break it down a little chunks, at least you get a win. You can get the end of the day and you're like, I did this little thing and it was progress. And now when I just you know, when you're getting that burnout, it's like you just need a win. You just like I need to feel like I'm not spinning my wheels and stuck in a rut. And I think that does a huge amount for you. It just gets that little bit going. And the other thing I'll try to say, I have found I have found that throwing my schedule at AI and I'll just say, like, does this make sense? Is this reasonable? And it does do a lot about saying like, hey, this is probably there's going to be a long day or the hour. I'll say like, here's what I want to achieve over the next two months. And then we'll say, yeah, you could do that, but it's going to be difficult or you should be able to do it. And it's like you said, you got to take it with a grain of salt, but it can help you get a little bit of a sanity check sometimes as well. I felt for all of these pieces. Yeah, I do use Gemini Enterprise to reschedule meetings for me. So I'll look at him to say, oh, can you email that person at that time reschedule and propose a new time and things like that? So you can you can do that for me for sure. So I do like that. Yeah, sorry, I interrupted you. No, I was just going to say I was actually going to dive in. So have you found yourself because now that we're sort of talking about it, and we've talked about needing some guardrails and stuff like that. Have you found yourself like dipping a toe in and edging more and more into AI and feeling more comfortable with it, or have you sort of just you have sort of like, here's a guard's rails and this is where its place is right now. This is its lane. I. Because I have a number of projects on the AI trust is at different levels for those customers or for those projects. For myself, I use AI a lot in my developer productivity. So that helps me a lot. Again, never get touch the code in production. Oh, it's very good for documentation as well. So I do use it to help me document things. I'll get it to produce Mermaid, which I use tools. I use tools for diagramming and things like that. So that's pretty cool. So that's really helped me become. No, no, no. Documentation's cut my documentation time in half. So that's great. You know, some of my customers, I have a banking customer in Europe. I've worked very, very closely with them to develop an in-house trustworthy, deterministic AI agent. And, you know, adoption, 85% of the companies using it or bank is using it with over half of the company using it daily. So, you know, that's that what I take. I would take that as a high success rate. And because it's got a deterministic wrapper around the tooling. So when you go from AI agent into tools, there is determinism in there. And, you know, we've got to build an AI control plane and doing some really cool stuff in there. And that bank is quite happy. Then I've got customers down the other end where the CEO accidentally left the AI recording on and a meeting, said some stuff that they didn't want recorded and it got distributed to everybody. And then it becomes public knowledge or whatever. So their trust in AI is very low, whereas, you know, I've got a highly regulated bank in the European system. And their trust is very high because, you know, it's it's had the data warehouse and the I am and the are back all set up properly. And everything's correct from the very beginning. So, again, opposite ends of the spectrum. And the customer down this end that accidentally recorded the wrong meeting with AI. The AI conversations dead in the water. It will never go anywhere. Whereas the other end, they're looking at other ways to do more with it. CEO has even talked about putting his own voice on it and stuff like that. So we try to find funny, but interesting. That's actually, you know, that's that's a funny thing. I don't think we've actually talked about that. We've not in a professional sense, but we've definitely had those conversations because of how we do recording and things like that for the podcast. And even some of the conversations that we've had, internal conversations and things like that we've had on the team, because we Michael and both we have a lot of remote people that we work with. And sometimes you get conversations, you get things on a call that, you know, you weren't expecting. And I'll just give a brief anecdote story as I had a developer that jumped into a Zoom call early one time and it was open and it was recording from the start because I just I don't trust myself to hit record often enough. And he was on there for about two or three minutes and was by himself and was working his way through a problem and was, you know, was sitting there coding while he was waiting and ended up sort of cussing quite a bit as he was like talking his way through the problem. And so I had to warn him. I was like, hey, next time you jump onto a call, make sure you mute yourself first. You know, there's those kinds of things. It's like, and I guess everybody's got those stories of like, somebody left a camera on, they didn't realize or somebody walked across the shot or things like that. But particularly now with the the AI agents that are out there that you don't always control that now as you'll have somebody will jump in. I've got a customer jumps in. There's like three AI agents that jump in that are note takers that jump in before he ever shows up. And it's just so you get all of these things. And it's just because people are trying them out and then they forget to shut them off. And, you know, you have to kick them out or block them and all that kind of stuff. But it's it's definitely, I think, a whole new world out there that we have different things we have to be aware of. It's like it's I guess it is. It's a it's now like that's just society, too, as people have got cameras everywhere. They've got recorders everywhere. AI seems to be everywhere. So you have to be just that much more diligent about about what you do. Now, I want to because we are sort of getting there on time, but I want to swing around to all the way back to the story she talked about. You're also a board member. Plus, you've been with a couple of companies is like, I think a lot of developers struggle with that. That bridge from being a developer to being somebody that's like on a board or even I think sometimes like a CTO, CIO, fractional kind of thing. So what would be and this is actually more of, I guess, even a little bit of a personal thing that's like, how would you address somebody that was that was, you know, mid level developers or something like that that's thinking like, maybe this is something I want to do, or they also maybe think that this is something that is never going to be for me. How is that? How does that fit in your engineer developer kind of mindset being in those positions? Yeah. One of the biggest joys I get out of my roles in life is giving a job to somebody or giving somebody opportunity to develop. You know, if I if I, you know, if I recruit you, I want to see you become the very best version that I can help you become in the time that I'm with you. And, you know, you're not going to I don't pretend you're going to be with my with me or my team forever, right? You're going to move on. And hopefully it's the next thing is bigger and better. And it's another step in your career. So for me, I'll try and work with these mid levels and identify where they want to go in their career and work with them. And I do spend a lot of time one on one with the people that want to. Some people don't want it. Some people do. And I'll try and work with them a lot. For my perspective. You know, in my 20s, I first started coding. I thought it was my entire role was nothing else, but knowing the tech inside out and being the best tech person in the room. And that's what I thought my role was. And maybe it was when I was like a junior or mid. You know, I really wanted to be the best. I want to be the person that could answer anything technical about the platform, about the language, anything. And I guess by the time I hit my 30s, I really realized that actually technology is all about people. So the time you spend working with people and understanding their problems, their needs and what they want out of the business or out of the software you're working on is really the connecting factor. Right. Businesses only exist to make sales. And the only way you can make sales is through people. All right. So for me, my advice would be focus on the technology. Absolutely. But also do not ignore the personal side of tech and really spend that time to connect with your customers, connect with your peers and connect with your management. You know, one of the first conversations I like to have with you know, any new customer or anybody that, you know, if I've gone in as a consultant or something, you know, what are your KPIs? What do you need to look good and to get your job done? You know, if I work with a new partner, I'll say, you know, how do you make money? What can I do to make sure that, you know, you get your reward? So, you know, when I work with a hyperscaler, for example, in their sales teams or their partner programs, try and work out what they need from me to deliver on their KPIs and not only my own KPIs. So I try and become cognizant of what my team members need as well that I'm working with. So I think that goes two ways and sort of communication for people that want to work up the ladder. You know, in team and larger scale teams, I work with, I tend to have a fork in the road for technical people do that, want to stay tech and focus on that. And, you know, that that takes you down, say, a tech lead role, an architect role, potentially an architect role is a bit special because there's a lot of business interpretation requirements, a lot of customer facing. But, you know, that's that's kind of one pathway. Or if you want to become more people, you know, that's the software development manager. That's that sort of thing. So, you know, I tend to see a bit of a natural fork in the road. If you're more of a people person, you might head down that path. Or if you're more technical, you head down the other path. Or there's the third path, which is the solo entrepreneur or the entrepreneurial part where you want to take a bit of both. And I'll try to think of a polite way of saying this, but I always said to myself, you know, I always work for somebody that I have problems with. I don't agree with their decisions necessarily. And if I was a boss, I'll do it this way. And I realized that for a long time that I use a more colorful language when I say to myself. And then I thought I should go out there and do this stuff myself, because if someone's going to make a mistake, I'd rather blame me and influence things myself than then sit there disgruntled about, you know, that technical decision or that product feature or that marketing campaign or whatever. So I'd rather be the person making the mistakes and working for somebody I think are making mistakes. So for me, it's about, you know, taking on the accountability and the responsibility and doing the things, those things myself. And I really love that fact that I can make a decision and make influence or make a strong influence on my customers outcomes or my software outcomes. And that's great. And of course, you know, I work for Fractional CTO roles and I help them, you know, deliver the best that they can in their product space as well. And, you know, even that gives me frustration sometimes. Like I've got some experiences and, you know, I've delivered some really large projects. My largest project had 80 million users. So that is a decent product. Seventy five thousand servers, if everyone's interested. Twenty thousand Cassandra nodes in three rings. But that all aside, you know, I've had some great experience on product development and in the consulting or the fractional roles. You don't necessarily get to bring that because there's somebody else's product vision that you're working with. And that kind of grates for me sometimes. But those are past the human things I talk about. You know, for me, it's like it's learning. I'll drop into an organization, learn something new and work out what I could do to improve things. And I love to practice that as well. So the advice is, is listen and learn, work with people and understand where the value comes from and where they drive the value. And you can apply that to your own projects and your products. And you really come much closer. To a great outcome. That's great. We are getting really close on time, so I just kind of want to follow up real quick. What are we've talked about on that and you've talked about using enterprise versus consumer models. What are some of the enterprise tools specifically that you use that our listeners could kind of check out? Because I'm kind of interested in what everyone's using because you just kind of hear the basics online. You got to dig into some of these other ones. What are some of the specific tools that you use that our listeners can check out? What are some of the specific ones that you like to use? Geez, I've used probably almost every paid for model that you can get out there. I think from Claude through to ChatGBT through to Gemini Enterprise. I've developed products with Bedrock. I've gone open source. I've done all sorts of things. My preferred tooling. I love Gemini Enterprise as a product because it hooks natively into my Google workspace. So I can expose Jira to it. It's got a whole bunch of in-built connectors that can automatically surface existing things I use. So I find that great. And then I've got their ADK, which allows me to write custom software and custom agents to do tooling. And that allows me to stick in the deterministic layer I've written. It allows me to stick in an AI control plane. And what I mean by an AI control plane, it monitors my tokens. So don't overspend. I've had a customer wrote an infinite loop to test their model out and they uploaded a document of 50,000 spaces or something. And they had like a hundred thousand dollar bill in 24 hours because a space is still a token or still adds towards a token. So even if you're sending up a blank document, that's still going to charge you. So, you know, these sorts of things to capture that, to protect PII, end-to-end encryption, all sorts of really cool stuff there that you should be doing. You know, I see, you know, even prompt injection synonymous with SQL injection that we tried to solve in the early 2000s. You know, people were injecting bad things into our databases. People are doing that today. So anyway, so I find those software development kits great. I completely avoid all consumer models. Totally. I never touch anything that's free. I never touch anything that's, I might try something as an anonymous user, but I'll certainly never put any real names, data or anything into it ever. And that's why I've leaned towards self-hosted open source LLMs with Alarm or other tools as well. So I tend to use those. Or Google Gemini Enterprise is where I wound up. So from my daily driver, I guess is where I find it comfortable. I find it very good because, yeah, I do a lot of Kubernetes work. I do a lot of Go as well, and they both come out of the Google stable. So, you know, Gemini is clearly trained on Google documentation. So it has a very close tie for a lot of my technical stuff. And because I use Workspace as well. So that's where I'm at. So I'm a bit skewed because of my Google reliance, I guess, in a professional manner. Personally, again, it's open source. I self-host my own servers and everything. I do my own email domains. Whitelisting my IPs are difficult as we all know for email serving. But yeah, there's a divide there and I keep them very, very separate. Well, I think we have, as Michael alluded to, we've pretty much run out of time. And once again, that is just like flown right by. There is, I think we could do this for a couple more hours and have no problem whatsoever. Just like lots. It's a lot of different questions and a lot of different things to unpack along the way. But appreciate so much your time. I want to be respectful of it. So one of the things is I know that you've brought up a lot of interesting things and cool ideas. What is the best way for somebody to get a hold of you? If they have some further questions or they just want to learn more about what you guys are working on. Yeah, what I've worked on recently, I've written a white paper on AI for execs and putting together strategies. So that's just to hit white paper dot download forward slash AI playbook. That'll get you to a document that I've collaborated with. It also has my LinkedIn details. I'm happy to hear from anyone on any topic. I don't outsource any of my data, so I won't. I'm not a good out provider from that perspective, but data is too too protected. I think Tim Berners-Lee said, you know, software comes and goes, but data stays forever. Whatever it is, I can't remember the exact quote, but data is where it's all out anyway. So hit white paper dot download. And then I'll just go ahead and go to the next question. Anyway, so I hit white paper dot download. There's an AI playbook there and it's got my contact details. And you can contact me without downloading the white paper if you like. But have a read. It's good for execs or people that want to put together an AI strategy. Excellent. Well, thank you so much for your time. And we'll make sure for those of you guys that are there, have that in the show notes so you can reach out and check that out. I think that I'm now curious about the white paper itself. I think that would probably be a really good read for everybody. You may want to like slip it under your boss's door or whatever as well. That goes or under his virtual door or her virtual door. Thank you so much for time. I appreciate you hanging out with us today. For all of you that are listening, thank you so much for again, for your time. Pencils down. You can stop taking notes for now. We will return next time with another interview with more information and just chugging away on how you can keep that forward momentum going. Until then, go out there and have yourself a great day, a great week, and we will talk to you next time. Most quality, reliability and support you can count on. Find out more at EnvisionQA.com. Thanks for tuning in to the Develop the Newer Podcast, where we're all about building better developers and better careers. I'd love to hear your thoughts or feedback. So drop a note to info at DeveloptheNewer.com. Be sure to subscribe on Apple Podcasts, YouTube, or wherever you listen. And remember, a little bit of effort every day adds up to a great success. Keep learning, keep growing, and we'll see you in the next episode.