Built from inside the culture. Not Silicon Valley looking in.

I didn't come to AI music from a software career. I came to it from real sessions, real studios, and real artists trying to finish records under real pressure.

My dad is a musician. My mom is an artist. I grew up in the San Fernando Valley on creative energy and not much else. If it weren't for my grandparents, we would've been homeless. They taught me how to work. My mom gave me the eye. That's how I ended up as a photographer, a creative director, a producer, and eventually the person Kanye West and Ty Dolla $ign brought in to design AI vocals for the Vultures records.

In 2013 I dropped out of film school after Ty Dolla $ign and DJ Dre Sinatra invited me on tour as their photographer and videographer. That's where I learned how artists actually work. Thirteen years later I'm still in it, and I've spent every one of those years either in the studio or building the tools to make studios easier.

The Vultures work

I was tapped in on AI vocal modeling before most of the industry knew it existed. Started making AI vocal covers of Juice WRLD and XXXTentacion back in 2019 and 2020, when RVC was a tiny, obscure technique that almost nobody could use without breaking it.

By the time the Vultures era came around, I'd been refining the workflow for years. I was around Ye and Ty for two years through the Vultures process. The first record was where the door cracked open. My own work inspired Carnival. The AI vocal layer landed on Vultures 2.

13 vocal models crafted for the project. Across multiple released tracks I served as Writer, Producer, and Vocal Producer. The work included AI vocal production, vocal execution, vocal reconstruction, and in several cases using my own personal vocals as the source material for AI generation, converting them into Ye's voice or layering them as new vocal elements throughout the records.

What I was doing wasn't replacement. It was craft. Writing, direction, references, takes, comping, cutting, editing, phrasing, taste. AI as one layer in a real vocal production workflow. The opposite of typing a prompt and pretending you made a record.

The Bully sessions

The year after Vultures, I kept working with Kanye's team on Bully. Same workflow. Sessions, vocal direction, AI vocal craft as one layer in real production.

There's additional unreleased session work from the Vultures and Bully period that isn't public yet. Vocal models, reconstructions, alternate versions. All of it sitting in my archive with full documentation. The receipts exist whether or not the credits ever get sorted out.

The wider session work

Before AI vocals, I'd already been working as a producer and creative director with Ty Dolla $ign, Wiz Khalifa, Don Toliver, Lil Pump, and others. Two production credits on Ty Dolla $ign's Featuring Ty Dolla $ign: "Lift Me Up" with Young Thug and Future, and "Nothing Like Your Exes." Photography and creative direction credits going back to a 2018 Revolt feature.

Currently still embedded with Taylor Gang as their E-Sports Coordinator and producing weekly episodes of the Tapped In Podcast. The day job has always been across photography, video, creative direction, and music. The night job is building Xport.

The pattern I keep seeing: the tools producers actually need are scattered across a dozen websites with credit traps, broken uploads, and unclear data policies. Real sessions break on bad tooling more than they break on bad ideas.

Why I'm building Xport Studio

I wanted to build this for a long time. The first idea was called Render Fusion. I hired two developers from Upwork to build it. One scammed me. One held my code hostage and demanded more money. Both were out of the country. I lost months and real money.

If you want something, go get it. Eventually people stop saying no, so why give up?

This is my third and final attempt. I'm building it myself now, learning the engineering as I go, shipping nearly every day. The architecture is local-first because that's the only way the tools can be truly trusted. Your audio never leaves your machine. Your voice models live on your disk. The engine source is auditable.

I'm not building this to look like every other AI audio company. I'm building it because the artists I came up around, the people I've actually worked with in real sessions, deserve tools that protect their work instead of monetizing every keystroke.

I don't chase. I grow relationships, have fun, and create. That's how I've operated for thirteen years and it's how I'm operating now.

What I believe

AI should fuel musicians, not replace them.

I've seen AI used at the highest level as part of a real creative process. I've also seen the cheap version. Empty. Soulless. Taste-free. The result of someone typing a prompt and calling it a song. Those are not the same thing. Xport exists to protect the difference.

The free tools build trust. The premium voice layer becomes the business. The marketplace gives creators ownership over the voices they train. The mission stays the same the whole way through: keep music people in the driver's seat as AI changes what's possible.

What you can expect from me

I prefer to let the work speak for itself and stay behind the scenes. That goes for the music I've worked on and it goes for this product too. The proof is in what ships, not in what I post.

I will ship. Xport is on its 27th alpha release and shipping fast. The work is visible. Bugs get fixed in days, not quarters.

I will tell the truth. When a competitor beats us on something, the comparison pages say so. When something breaks, I fix it publicly. When I don't know the answer to a hard problem yet, I say that too.

I will keep the trust commitment. Your audio never leaves your machine. Your voice models are yours. Your sessions are private by architecture, not by promise. If we ever violate that, you'll have the receipts. The engine is local, the telemetry is opt-in, every claim on this site is verifiable.