Michael Walker: Yeah. All right. Excited to be here with my new friend, Agent Method. So, what's up, man?

Agent Method: What's up, man? How are you?

Michael: Doing well! Let me give you a quick intro. So, Agent Method is a pioneer of the future punk genre. He blends pop punk, emo, and rock—which was my stomping ground growing up. I was an emo kid, so you're speaking my language.

His sonic innovation spans major artists like Illenium, Usher, and Lenny Kravitz. He's a sound designer, content creator, and he's gone from the DMV pop punk scene to global EDM stages.

Today, we're going to talk a little bit about sound design. In particular, I'm curious to hear your thoughts as a producer, engineer, and sound designer when it comes to the current state of technology for musicians—how they can express themselves in a world with AI and all these amazing tools. So, thanks for taking the time to be here today.

Agent Method: Thanks for having me. I appreciate it.

Michael: Absolutely. And I know right before we hopped on backstage, you mentioned that you woke up feeling a little under the weather today. So you were kind of debating—

Agent Method: —whether I could pull it off. So if I sound and look like shit, excuse me. We'll get through it.

Michael: Yeah. I give you huge props for that. We could have rescheduled, but you're like me. You were like, "Nope, I woke up today." How did you put it?

Agent Method: Big old cup of "get the fuck up and do it"—or something like that. I typed it to you; it's in your email.

Michael: It's good. And I would say, like, 99% of the battle—the thing that separates successful artists, or just people who accomplish meaningful goals—is that mindset right there. Just saying, "I'm going to make it happen."

Agent Method: You gotta start. I was watching this reel—it just came across my screen. I think it was actually an ad, not even a proper reel, but it was about this guy who created a methodology for motivation.

The idea boils down to this: within nine seconds of thinking about doing something, that's the prime time to begin. Your brain will start the "wanting to complete it" process because that's wired in there somewhere. But if you wait beyond nine seconds, it starts to turn into something else. I'm trying to adopt that more in my life.

Michael: Oh, that's some good stuff. It definitely seems like there are waves of energy or momentum. If you swim along with the wave when it's cresting, you can ride it and get all the momentum. But like you're saying, if you wait too long, the wave passes. And by the time you're ready to go after it, it's like—you kind of missed it.

Agent Method: Gotta wait till the next one.

Michael: Good stuff. Well, to start out, I'd love to hear more about you—your story and how you found your way into sound design and working with artists like these.

Agent Method: Yeah, man, for sure. I started off in pop punk and emo rock bands back in the day—middle school, high school—the DMV scene up there was pretty big with that.

Then I became the recording guy, working on my friends' stuff. A few years after high school, my buddies wanted to go down and work with producer Zach and Ken from ZK Productions.

Do you know who they are? Have you seen them?

Michael: Yeah! We've done an album with them.

Agent Method: Oh, awesome! Awesome. They're good buddies.

Michael: I said, man, we were definitely... same stopping.

Agent Method: Yeah, we're very well connected. Yeah, they're my homies. I've known them for over a decade. We went down to Treesound Studios in Georgia. I was still living in the DMV at the time, in Northern Virginia—Reston, Virginia, to be specific—and then came down here to work with them. It went really well. Tried to get an internship, but it just wouldn’t work at the time. Finished the project in about a month, went back up to Virginia, and then, a number of years later, went to audio school.

I have a bachelor's in audio production and MIDI and something else... I have a bunch of internationally accredited certifications. I'd have to go back and look. After that, the same group of kids wanted to go back down and work on a different album. So I went with them again, tried to get an internship, and they were like, "Why don't you just try to get one with Treesound?"

They had moved from Treesound to a different place, and I was like, "Yeah, okay, I'll go back there and try to do that." The project ended, I flew back up, called in, got an interview, flew down for the interview, and got the position. My aunt and uncle just happened to move to Alpharetta, which is just a little ways away.

I don’t know where you are right now, by the way. Where are you?

Michael: I'm in—

Agent Method: Orlando, gotcha. And they just happened to be moving down to that area, so I moved down with them for the better part of a year and started working at the studio. I had a bunch of clients there, and that got me in the rooms with Usher, David Banner, Lenny Kravitz, and a ton of other people.

I was engineering for them and doing sound design to a degree. What they would expect you to do, to some extent, is if you've got a track, sometimes the rest of it isn't built out, and you just kind of have to build it on the fly while they're recording. So there was some of that in there, but for the most part, it was just engineering.

Then I just kind of got disillusioned with studio work and went into the corporate world, which is a weird transition—to be disillusioned with the studio and go to corporate. I think it was just the consistency of income and security. I was the head engineer for Action Advertising for a number of years. They do car dealerships, furniture stores, and that kind of stuff.

Then I got over that pretty quickly, went back to freelance, and started working with clients all around the area. My wife suggested I start making some silly videos because I'm a pretty goofy dude, as some people may have seen from my videos.

Then, Steven Slate hit me up on Facebook because he saw one of the videos. We had been friends for a long time on Facebook from some long-ago connection. He was like, "I love these. Can you do these for our stuff?" So I did a bunch for them, and then he was like, "Do you want to do stuff with Anna too?"

I was like, "Yeah, absolutely." I do sound design all the time for myself. I don't sell it, I don't do anything with it. But for the past four, five, six years... I've been a sound designer for 20 years, but I've only been earning money from it in the past five years or so. And then it was off to the races from there.

That led to the next thing, to the next thing, to the next thing. I could use each as a stepping stone to show what I'd done in the past, and now it’s on the resume. Now I just have sound design jobs lined up. So, there you go.

Michael: Okay, cool.

Agent Method: Does that fill the box?

Michael: It definitely fills the box. Yep. And yeah, I mean, I'm curious to hear your perspective as a sound designer, as someone who kind of lives and breathes in this world. I feel like AI is sort of a buzzword right now, for good reasons—it's changing everything, especially in creative industries. There’s generative AI coming into the space.

So I'm curious to hear how you think about AI. Are you using it? How do you recommend people think about it as a tool?

Agent Method: Absolutely, man. I love coming up with ideas that I probably couldn't mathematically put together on my own because I'm not great at math. Feeding it that kind of input and having it output something that I could use in a song or in a patch for a synthesizer or something—it's incredible.

The easiest example would be Serum. It's got a formula parser, so if you can describe what you want in a formula parser, it can spit out an equation. You can put that equation in the formula parser and get a very cool wavetable that you wouldn’t have otherwise.

Virtual Riot did this thing, and I was doing it as well, where you tell AI to make a four-second wave file using only sine shapes—sine waves—and just do all prime frequencies, from the fundamental up to the third, seventh, and all the rest. It'll spit you out a WAV file like that. Then you can throw that in and do some wacky stuff with it.

Yeah, man, you can use it in really cool ways. And that's just scratching the surface.

In terms of practical applications, I’ve also trained some GPTs on manuals and stuff so that if I need to dive into something more deeply, and I don’t have the time to pour through the manual, I can just ask the computer, "Hey, you are the manual. How do I do this thing I’m trying to do?"

And then you learn something much quicker. You can apply that knowledge to something else. It’s really about expediting skill development.

Michael: That sounds like a lot of fun. So it sounds like AI allows you to sort of write formulas that you can copy and paste into certain tools to create sounds. You can describe what you want and have it turn code into music.

Agent Method: Yeah, effectively. Energy changing form, in a sense.

Michael: So, I’m wondering—because you also teach sound design to other artists—

Agent Method: Hmm, yeah, to some degree. Not a ton, but yes. I mean, yes, in the sense that I have videos that show that. I have a couple of private clients I've worked with over the years, but mostly my content is semi-educational. It’s mostly stupid, but you’ll learn something.

Michael: Okay. So I'm curious—when clients come to you, what are some of the biggest challenges they’re facing right now in terms of their sound design? What are the biggest problems they're trying to solve?

Agent Method: It's always like routing and understanding what's going to what and how that's affecting things. And getting your head around bipolar and polar but negative, and unipolar but negative, and flipped unipolar but negative, so you're going backwards—like setting all the modulations to go the way you want them to, based on what the parameters of the synth are.

And that's a lot to wrap your head around because it's all logic gates—1 plus 0 plus 0 plus 1 plus 0 plus 1—just going back and forth between that chain. And it was one of the harder things for me to grasp, especially with synthesis. And it seems to be reasonably difficult for the few clients that I have.

But then also, just creatively—what to do? Like, where do you start? That's another big one. And you only have nine seconds, remember? So it's just like—those are some of the challenging things. What to start with and understanding enough of the tools and how they interact with each other to have an idea and be able to manifest that idea.

Because you know what does what and how you can accomplish this tone or this growl or whatever. And I've learned a ton myself from YouTubers—like I said, Virtual Riot again, AU5, Virtual Riot, Dash Glitch. Just tons and tons of people. Steve Duda—not that he's a YouTuber, but he's on there from time to time.

Yeah, man, there's what to do and how to do it—always the biggest things.

Michael: I guess the "why" is sort of connected to the "what" as well, because you're trying to figure out what to do, and to figure that out, you kind of—what's your goal, or why?

Agent Method: That's very true as well. Yes, of course, and how it serves the song that you're writing or whatever.

Over the years, I've become increasingly more robotic in the way that I think. Mine becomes very much a sense of utility, and then other people that are better feelers than I am can really morph that well into emotion. Not that I can't—it's just I get more satisfaction out of the technicality, at least to this point in my life.

Michael: Cool. You're ahead of the curve, or we're all going to be robots pretty soon.

Agent Method: Yeah, right. Yeah, man. Think about that cyberpunk dystopian future. It's pretty wild.

Michael: Yeah, we interviewed Nolan Arbaugh on this podcast a couple of months ago, and he's the world's first Neuralink patient.

Agent Method: Oh.

Michael: Human cyborg. We created the first song ever telepathically.

Agent Method: I saw that. That was really cool.

Michael: It was cool. I mean, Nolan's awesome, so just hearing his story and seeing what he was able to do with his thoughts was remarkable. But it definitely makes me think about the future of all humanity and where things are heading.

Agent Method: Yeah, for sure.

Michael: AI and robotics.

Agent Method: It's a future we don’t know what's coming yet. It's probably gonna be pretty drastic. But we’ll see.

Michael: It definitely doesn’t feel like it's slowing down.

Agent Method: No, sure doesn't, man.

Yeah, man. In terms of the future of music, I see stuff online, and I would love to get into this more—I just haven't had the capacity for it at the moment. Just gestures and ways to make music—physicality.

Imogen Heap has really cool gloves that she and some people have developed to do that kind of stuff. And the Yamaha—something—not DX7, but GW-something workstation. It was a workstation synth/sampler thing. It had a D-Beam.

You remember what D-Beams are? Did you ever have a keyboard with a D-Beam?

Michael: I don't think I ever had a D-Beam.

Agent Method: It was an infrared and then a receiving light—bulb, whatever. LED, whatever it was. And just based on your position—kind of almost like a theremin—you could change parameters on that kind of stuff.

That was my first introduction, all those years ago, into movement into sound to one degree or another. And it wasn’t anything super drastic. It was like taking volumes up or down. I mean, you maybe could have mapped it to something different, but I wasn’t that heavy into sound design back then. I was still a fledgling audio kid.

I'll have to break it out and see if I can do anything cool with it at this point. It kind of reminds me of that.

Michael: Hopefully, it still works.

It's a cool idea, I mean, just thinking about mapping different properties or parameters to music and using that as a way to create music from sources you wouldn’t normally think of.

I mean, one property or parameter—it's like each of our fingers is pushing down at a certain point. But it does seem like with a neural interface, part of the implementation process of creating music using your thoughts will be mapping different parameters to different ways that you're thinking and then using those to express themselves through music.

Agent Method: Yeah, that's pretty wild. And forgive me, I can't remember—was he fully able-bodied and just did this, or was he disabled in some manner?

Michael: He is paraplegic.

Agent Method: That's right.

Michael: Yep.

Agent Method: So that was effectively the only way to—yeah, right. Wow.

Michael: Exactly. So, I mean, that's part of what makes it so remarkable—he's able to do this with his thoughts, doing something that extends the capabilities of what's possible.

Not just for someone who's paraplegic, but really for humans in general—just thinking things into existence. And in some ways, you could say that we always kind of think things into existence. It's just that our actions are an extension of our thoughts. But to directly have a thought and transmute it into matter...

Agent Method: Directly. It's pretty wild.

Michael: Yeah.

Agent Method: The future’s gonna be crazy, man.

Michael: Yeah.

Agent Method: Yes, it is. Wow.

Michael: So yeah, I'm curious—for all the artists who are listening or watching this right now, who are looking to explore sound design and really just expressing themselves in unique ways—what would your recommendation be?

How do they catch the waves that are happening right now? How do they stay on the cutting edge and avoid being left behind?

Agent Method: Yeah, totally. That's a good question. I haven’t given that much thought before.

Michael: I guess one way to do it is to tune into conversations like—

Agent Method: Sure, right. Yeah, exactly. The other thing I was going to say, which is an obvious thing, and it's kind of a cop-out answer—I don't want to give a cop-out answer—but it's very much the truth, is just f***ing YouTube, man. If you want to learn anything about what the functionality of an LFO does, just go to YouTube and just learn. Just do that thing.

Just do it along with it—screen on one side, the DAW on the other—and just get in there and start doing it one for one. That will start opening bigger and bigger doors for you faster than probably anything else. If you have a certain amount of creativity—which I'm assuming any person who wants to be in this scene does—you could think about all the rest of the things you could apply it to.

All the rest of the parameters of a synthesizer, or taking it out of the computer into using gear and stuff. I've got a UDO Super 6 synth right here. It's probably out of frame—it looks like it is—but it's an awesome synthesizer. It can do really cool things like using outboard inputs to trigger audio stuff, just voltage, effectively. All older analog synths were pretty much that.

But it's also fully digital with analog filters and stuff to a degree. So you don't have to, but just the functionality of, "Now I understand how to route this," I can think that far over here when I wasn't even thinking that far over here before. I was just twiddling with this one knob for two hours, trying to figure out what it did.

So it's just about getting out there and learning anything to start the train movement.

Michael: So it sounds like what you're saying is that really, the best way to learn something is to do it. It reminds me of Nike—just do it. But there's a lot of truth to that. That's the best way to learn—actually doing the thing.

And ideally, you can also learn from other people who've done it. Like you're saying, you can find someone on YouTube or follow along, but you've got to show up and do the thing. And even when it comes to learning, it's not enough to just learn—you actually have to actively learn, do the thing, get feedback, and iterate as you go.

Agent Method: And that's one of the nicer things about sound design. If you're following along with something and the sound you're getting matches what they're doing, you're on the right track. So you know that the decisions you're going to make from there on will probably be correct to reach the outcome you're trying to get to.

It's pretty quick, immediate feedback to say, "Hey, you're on the right path," whereas you don't always get that feedback. I can't think of another example where it's maybe not that fast, but I think you know what I'm saying.

Michael: Yeah, I totally do. One of the biggest challenges of trying to learn from a course or a video is that if you don't have that feedback, it can be a big limiting factor.

I know for me, here's an example: When I first wanted to learn how to work out and lift weights the right way, I downloaded some apps and watched some videos, but whenever I did it, I would always tweak something in my back or somewhere else. It wasn't until I got a personal trainer that he was like, "Oh, you're doing that wrong. You've got to use this posture and do it like this."

I was like, "Oh, that's really helpful!" And that's a good example—sometimes, you really need that feedback.

Agent Method: Instantaneously, yeah. If you're trying it on your own, exactly.

Funny enough, I was actually a personal trainer for like five years back in Northern Virginia.

Michael: Nice, what a coincidence!

Agent Method: So funny enough, I'm very familiar with that side of things as well.

Michael: Nice. Man, my personal trainer is definitely one of the best investments I've made in my personal health and life. And he's also just awesome—like, he's a good influence in general.

It's a good example of what you're describing in terms of that feedback. It sounds like what you're saying is that one of the nice things about sound design is that you kind of get that feedback. So when you're following along with something, you can actually listen to the sound next to the other one.

Agent Method: Mm-hmm, yep, for sure.

And from there, like you said earlier, it's iteration. It's like, "Okay, I don't want it to be exactly the same. I want to put in whatever functions serve the sound and what that sound is supposed to serve in the song."

So it really just primes you to try new things. Once you understand the instructions—like with the GPT bot thing I was talking about—once you've got the instructions, you know where to go, you know what the tools are, it's just about poking around and experimenting.

Michael: Yeah.

Well, what this reminds me of is recently, I've been on a learning-by-doing kick because I've realized that ChatGPT is sort of like the cheat code for learning a new software tool or learning something extremely quickly.

There's a shortcut on the Mac app—Alt + Space—and then ChatGPT is right there. One little hack for anyone watching this right now—if you're not already doing this—is that ChatGPT, while not perfect, is really good at keyboard shortcuts now.

So when I'm trying to learn Adobe Premiere, Blender, or any new software, if I don't know how to do something, I'll just ask it, like, "Is there a shortcut for this?" And it'll tell me, "Yep, it's just Command + E." And I'll do it, and it's like, "Wow, that's amazing!"

That actually makes everything so much faster. And there are tools I've been using for a long time where I'm realizing, "I really should have learned the keyboard shortcuts." I just never took the time, and now I see I can do things five times faster if I use those shortcuts.

Agent Method: Don't spend all that time on minute editing, absolutely.

Michael: Yeah, exactly.

Especially for anyone doing editing work, engineering work, or working with software tools—that's a great way to learn on the job, like you're talking about. Learning by doing, then just asking for help, whether through a tutorial or even just having ChatGPT next to you while you're learning.

Agent Method: Absolutely. I'll send you the link to the GPT I trained. Anytime you've got a manual, just toss it in there and ask your questions.

Michael: Awesome.

Agent Method: Yeah, man.

Michael: How off the deep end are you in terms of AI tools and the latest AI news? Is that something that excites you, or are you just tangentially following it?

Agent Method: Yeah. I don't really follow it so much. Sometimes it comes up in a feed on social media or something like that. I wouldn’t say I'm set in my ways by any means, but I'm just so involved in what I'm doing that I'm not actively searching for anything.

I just take things as they come by, and if it feels like it serves what I'm trying to do, I'll investigate it further. There was something I saw—oh yeah, the SynthGPT thing from Fader. Do you know about that? Have you tried that?

Michael: No, I haven't.

Agent Method: No, I haven't really tried it either. I'm interested to see how well it actually works because it's really just text-to-sound. It's kind of cutting out my job, but it's really just text-to-sound.

From what I heard, it's okay. I think Suno does a better job with patch creation in the sense that it has to create a patch to be a synthesizer or something. Maybe not create a patch—it just has to create the bits that sound like a patch, that sound like a synthesizer. So I guess it's not really the same thing, but...

Michael: I'm curious—you just mentioned Suno and some of the generative music tools. Is that a workflow you've played around with or one that you'd recommend to people?

Agent Method: Sure. It's a great source for just starting off an idea and then taking it however you want from there. It's amazing for that kind of stuff.

That's something I struggle with when I write anything, which I don't really do that much anymore. I kind of need to get back into that. It's cathartic. And I've just neglected that part of my catharsis for a while.

Having an idea really helps push you along. You can get excited really quickly—like, "Oh, I could do this here instead of this," or "I could change this chord progression and put it in this key instead." And it would be much more aligned with what I'm trying to accomplish.

Having it as a jumping-off point is far better than the nothing that was happening in my head prior to that. Or—not nothing, but the same old four to five chords you hit when you're playing the piano, the same things you strum when you're strumming the guitar. Just the muscle memory of it all.

Subverting that and taking things into new territory—if it's not something you usually do, there's probably some chance for growth in that.

Michael: Yeah, it definitely seems like a great jumping-off point and a way to get over writer's block or the "empty page" syndrome.

It's almost like having a co-writer in the room. It's nice when you can bounce ideas off something because otherwise, you can get stuck in your own head, overthinking the same thing or trying to tweak something that just doesn't work.

Agent Method: Right.

Michael: Really cool. My dad, just a day or two ago, was talking to a family friend named Gino. He had toe surgery, and my dad was like, "Hey, Michael, do you know the name of that AI music app? Could I do a remix of the 'Hammer Time' song, but change it to 'Hammer Toe'—like, 'Hammer Toe'?"

I was like, "I don't think it's that easy to do exactly that." I mean, maybe. Do you think you could literally recreate an AI-generated parody of a song? But I told him, "I bet Suno could do something pretty good."

Agent Method: Yeah, as a matter of fact, I was using it the other day to spark some ideas, and it played something that was almost exactly like Earth, Wind & Fire or something. The melody was so close.

So yeah, you could definitely give it a direction to go—no doubt.

Michael: Well, that's what I ended up doing. I used Suno, copied and pasted what my dad asked, and it turned into a really great song. But it definitely wasn't that close to "Hammer Time"—it was kind of a different thing altogether.

Agent Method: Yeah, and you can always tell when it's Suno or AI or something because it has this weird, almost unattainable, kind of chorus-y effect on the vocal. It's just the result of bits that aren't the vocal being around the vocal.

I don't even know how to explain it, but it feels like the tightest vocal stacks—just lots of them, all the time, with AI-generated stuff. It has yet to come back really, really clean to where it's super believable.

Michael: I feel like I heard the same thing. I think Udio.com—that AI tool—sounds more realistic to me. Some of the things I hear from that one make me go, "Oh my gosh, I wouldn't have known that was AI."

Suno, to me—especially Version 4, which came out somewhat recently at the time of recording this—is fun and remarkable.

Agent Method: Much better than Version 3, for sure.

Michael: Yeah, but it still has that AI sound to it.

Agent Method: Which is an interesting effect. It's probably not just stacks of really tight vocals on top of each other—unless it's building on an iteration. Maybe that's it.

Maybe it's just one version of itself, and then it builds on it, and builds on it, and builds on it, and you're hearing all the iterations at once. Maybe that's why it sounds like that.

I have no idea. I've never programmed AI before. I've trained them, but not programmed them. Interesting.

Michael: It is interesting. Well, Agent, man, it's been a lot of fun talking today—geeking out a little bit about AI and the current state of things.

I'm glad you decided to wake up today and just do it.

Agent Method: For sure.

Michael: I appreciate you taking the time to hop on the podcast.

For anyone watching or listening who’s interested in connecting with you or potentially working with you in terms of sound design, what's the best place for them to go?

Agent Method: Totally. AgentMethod.net. And my YouTube, but you can probably click through those links on my website. It's really just AgentMethod.net.

At the bottom of the page, there's a "Contact Me" form that sends messages directly to me. There’s also some free stuff on there for people to download and mess around with—videos to watch, songs to listen to—all that stuff.

Michael: Awesome. As always, we'll put all the links in the show notes for easy access.

Thanks again for taking the time to be on the podcast.

Agent Method: Thanks for having me, man.

Michael: YEAAAH!