AI Tooling Doesn't Mean You Skip The Engineering

Transcript is auto-generated and may contain errors.

Hey folks, I'm just driving to CrossFit here. We're going to the comments. This one's from I believe it's pronounced A Reacts. Um they left a comment after stumbling upon the channel and they're Oh my god, this car has got to stop. I'm going to lose my mind. Their question was around uh getting into the software industry. and whether or not I don't know like AI and the landscape that we currently have is is kind of destroyed software engineering as a as an opportunity and and my thoughts on that. So, I thought this this would be good to talk through cuz I think it's not a not a super I don't know unique experience in the in the sense that like I think a lot of people are kind of struggling with this kind of thing. And and I I don't know.

I guess I I feel like to me it seems like there's there's a lot of reason or a lot of things going on that I I feel like give give reason to uh you know people having concerns and I get that. And then I also think there's like this sort of uh I don't want to call it like a gap to like suggest that people like aren't intelligent. That's not my not my intention, but like uh there's like a gap in understanding I guess. Um and I I feel like that gap is something that would like help to understand especially when I'm answering questions like you know do I think that there's like the software engineering will cease to be kind of thing. Um, so this, you know, this person is saying that, you know, the things that they really like are some of these these challenges where you're really like debugging things, you're actually programming, like you're you're doing the hard thinking, right?

Like this is a pretty um I would say pretty typical sort of situation where someone is very curious. They want to understand how things work. They want to like solve the hard problems. And if you're if you're moving more things towards AI, does it start to feel more and more like well something else is solving the problem? I'm just like kind of like I'm just here typing prompts and not really thinking. And I guess like I'm hoping that by the time I finish this drive to CrossFit, I'll figure out a good way to say this. But my my first attempt at this is kind of like I I I feel like that uh we're being given a tool or sets of tools. So we have AI and it can write code, it can read code, it can do things a lot faster and in more volume than we ever could before.

And so it's like great, now that we have this tool, like doesn't that mean that uh like I don't have to like I don't know like be curious or like solve hard problems cuz all that I have to do is figure out the right prompt for AI. And it's like I I think that's the wrong framing because that that suggests that just because you have a tool that can navigate through things like this means that you must use it in ways that are it's it's almost like taking a a hammer to to screw something in or a screwdriver to hammer in a nail. Like can you use AI to go investigate problems? Absolutely. Can you use AI to go debug things? Absolutely. Can you use AI to help you uh with architecture and design? Like absolutely.

But does it mean that any time that you ever need to think about debugging something that like that you must immediately go to clawed code or co-pilot CLI, whatever you use, and you have you have no choice but like the only way that you can move forward is to to type problems. And the answer is no. And I'm hoping that when I say it like that, it sounds ridiculous because you'd be like, "Well, obviously obviously not all of the time." And that's where I want to be like, "Exactly." Like, you know, it's it is yet another tool. Yes, it is a very multi-purpose tool. Yes, it is, you know, able to do some very powerful things and very quickly and large volumes. Yes, of course. All these things awesome. Um, does that mean that like you are forced to use it in every situation from the start and you're not allowed to think creatively or like navigate difficult problems?

No. And so I think that like my take on this is like there's always going to be these types of problems to solve. Always. And I think for those of you that are very interested in in this part of engineering, regardless of whether it's software or not, right, like just this idea of of engineering, which is, you know, being curious, solving hard problems, understanding how things work, pros and cons, analysis, engineering, right? If you are the kind of person that enjoys doing this, there there will always be like more of this kind of work in general to be done than there are things to solve it. And so I feel personally that when we have things like AI to help us move through lots of problems, move through them quicker, move through more challenging problems, like that's great, but there will always be more problems to solve.

Always. There will always be some things that are like, oh, AI is not doing a good job of this. or AI is giving me conflicting results. I have to go understand this. There's always going to be some things like this. I I mean I could I could go on in terms of like philosophy on this, but I think my personal take is there will always be more stuff. It's like the other uh the other day I made a video on on measuring productivity with AI, right? It's like, you know, does it mean because your your company uses AI for everything that all of a sudden you've you've run out of problems to solve? Like, no. There there's always going to be more work. Always.

And I do think that what that ends up meaning is that it's almost the opposite, which is going to sound funny, almost the opposite thing I imagine happens, the the better AI gets, the more we're using it, the better we get with using it, the more that the easy problems are solved, and the more leftover hard problems you have to deal with. Now, does that mean that, you know, I see a future where you'll you'll never be prompting AI? Like, no. I like I think that's I think that's going to be part of the role, at least for the foreseeable future, until someone figures out like how it how it mostly does all the stuff that you normally prompt it for automatically, right? Until it's reading your mind kind of thing.

Uh but you know until that day comes, yeah, you're going to be chatting with AI and like it would be the same almost to the point where you're chatting with people on your team, but I don't think that means that like the the hard problems go away. I think there's going to be hard problems left. So when someone writes in like this and then you know they're kind of asking at the end like hey like is the industry done for my take is absolutely not. I think that there's there will remain tons of opportunity. When I say tons of opportunity I'm not saying that like oh I'm you know this this Nick guy on the internet saying well the job market's going to get super easy. No that's not what I'm saying. I'm saying there's tons of opportunity to solve problems in software engineering tons and I I also think that the more advanced technology gets also the more interesting difficult sets of problems there are to solve.

So, I think when I say opportunity, I'm suggesting there's always going to be more hard, interesting problems to solve. I don't know what the job market's going to look like for the foreseeable future. All that we know is that in recent time, obviously, it's been pretty brutal. I think there will continue to be some companies that have this idea that um you know I think some some companies that are like hey we can get rid of people replace them with AI some uh some companies that are like well we can't afford the people that we have so like we'll downsize to use AI and hopefully get you know the same productivity. Um I I think we'll continue to see this for whatever period of time. I I don't think that it is a sustainable way to run a business.

Like I I have not in the last 5 years that I've been at Microsoft, I've never I've never felt like, oh, if only I could just have more AI, I could get more things done with my team. Like absolutely not. I need to give responsibility to people so that they're empowered to do work, make very informed decisions as engineers. And I need more people like that like that. Am I I have not changed my mind on that. I I can't I honestly can't imagine being in a position where just like, oh, just add add one more agent. Like my dude, if I had no people on my team and I just had a bunch of agents, this would be like how I program at home, it would be a disaster for managing a team for a whole service. Maybe in the future that's different, but oh my goodness, we are not there.

And I say that as someone who codes a whole bunch with AI at home. I could not manage a team of just agents. It would be insane. So no. Do I think the industry is cooked and it's never like No, I don't think that. I think that it is uh you know forever impacted and changed of course but I I can't comment on the hiring part because I just I foresee that for some period of time we're going to have companies doing stupid things. I think we're in a bit of a turbulent time and I think that that will change. Okay, my camera's been totally lopsided the whole time. So, enjoy. Um, now what do I recommend people like this do? Uh, I would say continue to focus on fundamentals, continue to focus on areas that you're interested in, right? Continue to explore things.

The whole whole point right is if we think back to what I was saying about what how do I perceive what engineering is being curious, understanding how things work, solving hard problems, doing analysis, right? And being able to back up pros and cons like that so you can make good decisions. To me, this is the engineering part and you know, can you use AI to help you with that? Of course. But at the end of the day, like that's the part that we're doing. So, I think that it's important that especially before you're in the industry as much as possible, exploring different things, trying them out, right? Um I don't tell the story too often of like going to going to university like making the decision to go what to go into but I went for uh computer engineering at the University of Wateroo and the reason I did that is because uh I really liked programming but my my father used to run a a software company.

Well, sorry. He used to run a the whole point is that it wasn't just software, software and hardware company. He had a a technology company and as a kid he used to tell me no the the money is in hardware. The money is in hardware. Like if when you you know I can't I can't encourage you enough to move into hardware. And so I was like hey like I think this is a good balance if I go for computer engineering. Number one I know there's programming in that. Number two, uh I wanted to be an engineer. Uh and why I think I had a lot of people telling me as a kid like go be an engineer because of like uh like because of your interest and how like you like you know investigating difficult things and understanding them and being curious like go go into engineering.

So these were the things that influenced me to do that. But the point is that when I got into university and I was learning about the hardware side of things, I was like, "Oh my god, I hate this." Like I I want nothing to do with this at all. And so my my opportunities in university for what to learn about my I had six internships during university. So I had two full years of work experience by the time I graduated because of internships. that allowed me to explore different things. So I I got to do embedded software development for uh for two semesters. That was eight months. So I got to see like here's being close to hardware. Like I'm not the one like actually building circuits but I'm putting code onto boards, right? I got to work in startups. I got to work doing medical stuff, doing uh you know bank related things.

I got to work in mobile apps. I got to explore a bunch of stuff and I like I can't encourage that enough. Buddy, you have to hit the brakes when you're literally driving into the intersection. So, yeah, take take this opportunity. It's not that you can't explore things later. I would also encourage you at any point in your life, you know, you are an engineer, right? you if you feel destined to be an engineer, you're going to be curious and want to learn things and understand how they work. Like don't let that ever go away. Um I'm just saying like before you're actually, you know, working a full-time job in a particular space, like take the opportunity now, as early as you can to to kind of get that momentum exploring different things. And so that's what I would recommend is like I I'm assuming some people are probably hoping to hear like go take these exact courses, do this exact step-by-step plan, but like that's not doesn't make sense.

That's not advice I could give generally speaking to an audience of people. My advice generally speaking to an audience of people is to continue to explore different things. See what you like and what you're interested in because at the end of the day when it comes to working like I really hope that you find something that you are passionate about. Not saying that's easy especially like you know what company you work for the domain you're in. Some people never find that in their career. And I'm not saying it's impossible. I'm just saying like I think it's it's not trivial for people to do. So I think exploration's really important and I think that in any of these industries you're going to find opportunities to be solving difficult problems that you know you're not one one shot prompting AI to go solve everything for you and sitting back.

Will you have some of those? Yes. And that's going to let you go work on the harder which to me is good news. So, I hope that's kind of helpful in some way. Um, I don't know. I I I keep getting this feeling that there's more and more people that are having a hard time getting jobs and then they watch some of these video responses and they're like, "No, it's bullshit." you know, there's there's no chance no one's hiring. And I'm like, I don't I just don't know what to tell you, man. Like, I had to respond to someone the other day and be like, I've literally I've literally hired people, and if I could hire more, I absolutely would be adding more people to my team. We'll do that. So, I I realize it's not easy right now, so I'm sorry about that. But yeah, that's what I got today.

Thanks for watching, folks. Um, if you got questions, leave them below in the comments. And otherwise, I will see you in the next video. Take care.