In two lines

BAYSGATE is a songwriting-led music project.
When Code Learned to Feel is a 29-track album built from human writing, shaped with modern tools, and unconcerned with staying in one genre.

That is the long story in two lines.

What this is
The project

Writing first. Everything else after.

BAYSGATE starts with songs, not prompts. A line, a mood, a title, a melody fragment, or one stubborn feeling that refuses to leave quietly. Some tracks arrived quickly. Others took their time.

The record moves around on purpose. It never wanted to stay in one lane for long, and forcing it to would have made it smaller than it needed to be.

Why it exists

Giving old notes a future.

Part of this project came from a simple question: what happens to all the writing that never quite finds its form? Over time there were notebooks, fragments, rough lyrics, half-finished demos, voice memos, and pieces not yet songs.

BAYSGATE became the place where those pieces could finally become finished. Not as nostalgia or housekeeping. More like recovery.

The sound

Not one aesthetic. Not one mood. Better that way.

This is not a neat, single-genre record. Across the album there are traces of pop, alt-rock, retro songwriting, slow-burn romance, theatrical shimmer, instrumental drift, darker turns, and the occasional song that simply refused to behave itself.

Some tracks are bright and immediate. Some are stranger or more intimate. Some lean modern. Others reach backwards. What holds them together is voice, feeling, and intent.

pop alt-rock retro glow slow burns left turns beautiful mistakes
How AI fits into it
The method

A tool, not the source.

AI is part of the process, but it is not the origin of the work. The lyrics are mine. The direction is mine. The editing, shaping, rejecting, rewriting, and knowing when something still is not right — that is mine too.

What changed was the ability to hear things that would otherwise have stayed trapped in text or half-finished ideas. It made it possible to test voices, moods, arrangements, and versions much faster than I could have reached alone.

Why it still counts

It is real because the choices are real.

A song does not become meaningful because the workflow looked traditional from the outside. It becomes meaningful because someone meant it, shaped it, cut it apart, rebuilt it, and knew when it was finally saying the right thing.

I am not very interested in tidy categories like authentic or fake, human or machine. They are too simple for what art is, and too simple for what making things feels like.

What matters here is not whether the tools are modern. It is whether the song lands.

This is not a novelty project or a shortcut dressed up as a record. The hard part of songwriting never went away. The tools changed. The standard did not.

AURAVox
The turning point

The voice that made the project feel like a record.

A big part of the project clicked into place when the vocal identity finally felt right. Not just polished. Not just impressive. Right. That was when the songs stopped feeling like experiments and started feeling like they belonged to the same record.

The technical side mattered, but less than people might think. What mattered was that the performance finally carried the songs with enough conviction to let them just be songs.

prompt / direction lyrics vocal reference takes painful decisions final performance
Making the album
2017–2019

Notebook season

Lyrics, fragments, cheap headphones, late-night DAW sessions, and no serious album plan yet. Just a growing pile of ideas that kept pretending not to be songs.

2020–2021

Pressure and intent

The songs started talking to each other. The bigger question arrived too: what happens to all this if it never gets finished? That was the point the project stopped feeling optional.

2022

False starts everywhere

Hundreds of demos, wrong voices, wrong timing, wrong budgets, and plenty of almosts. The material was there. The right way to carry it still was not.

2023

The breakthrough

AURAVox changed the pace of everything. For the first time, the vocals felt believable enough that the record could stop being theoretical and start becoming real.

2024

Brutal edit season

Tracks were rebuilt, cut down, rearranged, and pushed into shape. Hooks moved. Bridges vanished. Better versions won. The album started to behave like a finished record instead of a folder full of possibilities.

2025

Release window

The songs stopped living on hard drives and started living in other people's headphones. Which is the part where all of this finally becomes public and slightly more frightening.

What holds it together
Not genre

Range was part of the point.

Different songs needed different shapes, so the album was never meant to sound tidy.

Not technology

The tools helped. They are not the thesis.

The story is still about writing, feeling, taste, and which version was worth keeping.

The thread

Voice, memory, and instinct.

That is what gives the whole thing continuity, even when the songs are doing very different things.

Why the name

BAYSGATE sounded like a place. That helped.

The name fits a project built around thresholds: between old writing and finished songs, private notes and public release, human intent and new tools, and something imagined for years and finally heard in full.

In that sense, BAYSGATE is less a concept and more a location. The place where unfinished things finally got through the door.

Go deeper

If you've made it this far, you're probably the right kind of nosy.

When Code Learned to Feel is the first full-length release from the BAYSGATE project. If you want the songs themselves, the Tracklist page has the lyrics and notes. If you want to hear what is already out in the world, the Releases page is the sensible next stop.