3 Pieces of Advice for Developers in 2026

Transcript is auto-generated and may contain errors.

All right, from the experienced dev subreddit, this post was three pieces of advice for new developers as we enter 2026. Whole bunch of New Year's themed going on cuz uh it's a new year. So, let's chat through it. I got one idea and I'm hoping that I have a couple more as I start blabbing away here. If you're new to the channel, this channel is driven by your questions. So, if you got them, leave them in the comments. You can go to code.com, submit questions anonymously there. If I don't have any that are waiting, then I just go to experienced dev subreddit or I blab about other stuff. So just send in what you got and I'll do my best to answer questions on software engineering and career stuff. So three pieces of advice for new devs. Um first one, uh you know, one's going to be about AI, so I'm going to leave that to second.

Uh first one I want to remind folks that uh software engineering is not a it's not a career path that you're going to excel at if you're trying to avoid people. Okay. So most software developers I don't have stats to prove it to you. But um unless you're doing like freelance work even then you're going to be working with people. But uh most software is built in teams. And so I think you know stereotypically there's a lot of us software developers that we love computers. We like being in our own little world building stuff. And I get it cuz I am that person as well. Uh truly right. I'm extremely introverted. I love, you know, lock myself in a room. Let me build cool stuff. If I'm not building stuff, I like to play role playing games, like hack and slash role playing games and just like get in the zone, right?

Like I like doing my thing, being focused on that. But um truly building software in a professional setting is going to be in Teams. And I cannot stress this enough that you need to make sure that you're comfortable working with others. You're comfortable communicating ideas with others. You're um you know able to give feedback. You're able to receive feedback. You need to be an effective team member. And that means that you know you got to communicate stuff clearly. It means that you can't just think about yourself when you're building things, right? You can't just get lost in the lost in the sauce, right? You can't just be hyperfocused on code where, you know, you're spending a week refactoring something, you come up for error and you're like, "Okay, now that it's done, it's still not perfect. I want to do it again." Right? like there's other stuff going on that's not just code.

And um I just want to remind people of that, right? Especially if you're a newer developer. You know, the the projects that you work on as a hobby, those are great. I think they're important to do, the school assignments you get. Um great to make, you know, to deliver on those, to to work through them and learn things. Like, don't get me wrong, I think all of that's very important. building things is important. Um but but please don't forget that uh you're going to be building with other people. And uh one more sort of point on this is when I say building with other people, I don't just mean other people writing code, of course, right? You're going to be working in a code base together with other people. There's going to be people that, you know, depending where you're working, they were there months, years, over a decade before you in a code base.

There's going to be some you'll see and you'll go, "I thought we weren't supposed to do that. They told us in school we weren't supposed to." And you're going to see it and there's going to be lots of it. And that's okay. It's not going to be perfect. You're going to be working with other people in the codebase, but you're also going to be working with tons of other people that aren't just programmers, okay? You're going to be working with product managers who are going to be helping shape the features and the requirements for the software based on what a customer needs. Okay, that's their entire focus is trying to help frame what what are the challenges that users are facing and they're going to be working with you, other engineers, your engineering manager, project managers. They're going to be working with all of you to try and make sure that the development teams can understand the pain points of the customer.

Okay? You're going to be working with people like that. You're going to be working with project managers. Sometimes you're going to have project managers and uh it's going to feel like they're micromanaging you or they're they're always asking for status updates, but it's not because they don't trust you, right? It's not because they suck. It's because part of their job is being able to make sure that things are on track. So, you're going to want to work with these people. They're not your enemy, right? You're all on the same team. You're going to have an engineering manager. You're going to have team leads, tech leads. If you're new to this channel, one thing that you will hear me say over and over and over again is level set expectations with your manager. This is one of the most important working relationships for you to establish. That's all around communication.

Okay? So, uh, I urge you if you're like, I don't like talking to people or like you it's it's uncomfortable or you avoid it or whatever it happens to be, this is one area. Please invest time into this in 2026 and beyond. Don't run away from it. We don't get better at that we avoid. We get better at avoiding it, right? So, um, so don't, like I said, I'm an I'm very much an introvert. It takes a lot of energy for me to uh to talk with other people and that makes it really difficult as an engineering manager, right? My entire job is talking with other people all day every day. It takes a lot of energy out of me. So I, you know, believe me, I understand what that's like, how that can feel exhausting, but it's an important part of your job. Okay? So I want to remind you of that.

Um, again, I should probably mention to you if you're new to the channel, if you're like, I don't even know who this guy is, and why should I give a about what you're saying? Uh, very fair question. I'm just a dude, right? I'm one guy. Uh, but I've been an engineering manager for 13 years now. I've worked at startups. I worked at a startup that got sold for uh went public and then got bought back to private for $2 billion. Uh, unfortunately, I didn't get any of that money. That's okay. Um, but they did pretty well. And then I work at Microsoft now as an engineering manager. Um been doing that for about five and a half years. So I've I've seen some First eight years as an engineering manager. I was a technical engineering manager. So I was also building the software with my teams.

And uh I program every day outside of work cuz I love to build And I've been doing that for you know programming for like 23 years now. So, um, I don't want to ever make you feel like I'm trying to tell you this is the way things are. This is the only way. These are just my my experiences, my perspective. Okay, that's point number one, communication. Invest time and effort into that. Okay, it's critical. Number two, AI. Yes, AI. You're all sick of it. I know. Um, AI is not replacing our jobs, okay? For the last year or two, they've been telling us, "6 months from now, you're all replaced." Um, we're still here. 6 months from now, we're also going to be here. It's okay. Uh, there's companies that have been making stupid mistakes like letting go of their dev teams because AI um and yeah, stupid mistakes.

AI is absolutely getting better. The tooling is improving. The models are getting better. Uh, the last video I made, some of the things I was talking about are just really how much things have improved. And so I encourage all of you don't shy away from AI tooling, right? I I really do think you should be using it. But I think that you should try to learn your tools. Learn where the strengths are. Learn how to utilize them. Learn where the weaknesses are and spend time with them. It's it is just a tool, right? It's a very powerful tool, but it's like a debugger, right? And not the same tool obviously, but you know, a debugger is a powerful thing. If you want to use a debugger effectively, you have to work with it, right? You can't just say, "Okay, well, breakpoint and like debugger go burr." Like, no.

Um, you have to understand like how to use it effectively, right? You have to understand how to use your AI tools effectively. We have lots of them. I think the only thing that we get more of uh than JavaScript frameworks is AI tools, right? Every it feels like every minute there's some new groundbreaking AI tool. Oh my god. I know we're all sick of it. But um I I do think that it's important that you you can't go use every single one of these things. It would be a waste of time. And by the time you're familiar with one, there's a hundred others. So spend time with them. Don't try to use them all. And learn them, right? Learn how to use the tools effectively. Look for the patterns. Try to understand how to get the right context given to an LLM. Understand how to give guard rails to your agents to do work, right?

How do we prompt agents effectively? How do we keep them on track? How do we prevent them from making the same mistakes over and over? Are you looking at the mistakes your agents are making, trying to refine your process over time and put better guard rails and prompts in place so that they don't make the same types of mistakes over and over, right? This iterative cycle. So, I I do want to encourage people to be leveraging AI tools because uh while I do think from like a a financial perspective, there's probably a pretty big bubble going on, I don't think that the the tooling is not going away. It's only going to keep getting better. And I I honestly think that, you know, the better that you are as a software developer in general, the more effectively you can lean into using some of these tools to amplify your abilities.

With that said, if you are a new developer, I also encourage you to not um you know, as much as possible, don't just blindly take AI output. Um even those of us that are experienced developers are not doing that. because uh just like people makes mistakes, right? I'm not going to let random people push code into my codebase and not review it, not try to refine it and prove it together. So, I do encourage you review the AI code if you don't understand what it's doing, right? If you don't, AI is a tool that can help you understand that. So, instead of going, I don't know what that means, man. and like let's just accept it. Switch your your LLM into ask mode instead of agent mode and ask it why what does this pattern do? Why are we doing it? Why are using this versus that?

Why did you pick this package versus this other one? Ask it. The best part is that you can ask it to explain to you in ways that you might be able to understand better. It's not like trying to find the right keywords to search for a blog article or a YouTube video, right? I say this as someone that makes both of those things. You can ask an LLM to explain concepts to you in ways that you might understand better. So, please take the time, ask it questions, learn from it. One final note on that is cross-check it. Don't just assume that it's telling you the right I work with LLMs all the time to build features in my social media posting platform, Brand Ghost. I do a lot of feature development with LLMs and I do a lot of conversations going, "Hey, does this platform support X, Y, or Z?" And it'll go, "Nope, nope, definitely doesn't do that." And it's like citing internet sources.

And I'm like, "Okay, but like uh I was telling my wife the other day, I said, "I want you to try harder." I literally told this to chat GPT. I said, "I want you to try harder." and it searched again and it proved itself wrong from its initial research. How ridiculous is that? A frustrated response for me which was try harder and then it did somehow. Right? So like don't just blindly accept things, challenge things. This is a good thing to be doing in software engineering in general except when you're working with an LLM, you can be an to it. when you're working with people, you can't. Okay, so that comes back to good, clear communication and being someone that other people want to work with. Okay, that's two. We got one more. Um, I would say especially for newer folks getting into development, I'm going to go uh a little bit uh step back from the technical stuff and this is more of like a sort of career kind of thing is network.

I highly encourage all of you to network. Okay. Um if you're not currently employed right now, uh you probably have noticed that there it's really challenging for getting jobs right now. There's a lot of things that are exacerbating that. And I think you know if you look at job applications that go up, there's like hundreds and hundreds and hundreds of applicants in like minutes. like it's it's kind of insane. Um, and I don't want that to to be so discouraging for you that you're like, "Well, I guess I better give up on software development." No. Um, some of these things just feel kind of broken now because like people can, you know, bulk apply to stuff with, you know, they're not even reading what the job description is. It's kind of like I'm taking a shotgun and blasting everything. Eventually, something's going to stick. But so many people are doing this that it's like makes it kind of ridiculous.

So I do highly encourage you to network. Okay, I am at Microsoft right now. I've switched teams at Microsoft. The reason I'm on the current team that I'm on right now is from networking. Okay. I make courses that I sell online and I've been able I've been very fortunate to be included on one of the most popular C course creation platforms called dome train that's from networking. I have been uh you know trying to interview software developers and aspiring software developers and this was total coincidence. I met someone at a conference I invited her on to my podcast, right? Trying to talk about um she's uh works from like a psychology background. So, we were talking about how that overlaps with software development. She ended up getting recruited because someone watched the podcast. I'm not telling you this like my podcast is like the Joe Rogan Experience or something, right?

It's not it's not like that. It's it's quite small, especially in comparison to that. and just from networking and having these opportunities, she was able to get someone uh that was interested in hiring her. Right? So networking is not a silver bullet. It's not when I say networking, I've tried to make this very clear to people in previous videos. Networking does not mean let me go on LinkedIn and spam my resume to 200 people until LinkedIn blocks me. Um, no, that's not networking. Networking takes time. You need to be intentional about it and uh you might not notice like anything tangible for a while, right? In reality is you might not ever notice anything from it. Perhaps at least from like a job placement perspective, it might not equate to anything. But I do honestly believe and this is why I'm encouraging it. I do honestly believe that it increases the surface area for luck or opportunities.

That's why I shared those examples with you because those are all things that never would have happened without networking. It's not a guarantee by any means, but those happened, right? I do obviously I do content creation. and I'm filming a video in my car. Um, but like I there's a couple of uh, you know, C content creators that I'm going to be meet. I've met them last year in person, which is awesome because they're from Europe and they're coming back to the United States in uh, the beginning of this year and I'm going to meet up with them again and they're bringing another guy and I'm super pumped to meet him too. And again, this is all just from networking. I cannot stress enough that I I encourage you find ways to network. Go to meetups, you know, meet people in online communities, go talk with other people doing software development.

You might find that it doesn't result in a job placement, but you might be able to build really cool meet really awesome people, and like I don't know, man. Like, that seems like a seems like a win to me. Okay, so those are three things I recommend you spend some time on 2026. I hope you found that helpful. And again, if you have questions about anything software engineering, leave them in the comments or go to code.com, submit a question totally anonymously or just look for dev leader on social media and send me a message and I'm happy to try and help. See you in the next one. Take care.