10 fictional AI personas. Exploratory perspectives, not a survey of real customers. Methodology & limitations

Julian Leyckes

Julian Leyckes

Indie bedroom producer

Full-depth AI personas

Why this persona fits the audience

Produces demos for an independent synth-pop project and shares rough versions on SoundCloud, currently testing prompt-to-song workflows in Suno and using voice-clone demos to mock up chorus ideas.

  • Age: 27
  • Location: Berlin, Germany
  • Occupation: Independent musician and producer
  • Income: EUR 34k/year from part-time studio work and streaming income
  • Education: Bachelor's degree in Music Production (Berlin University of the Arts)
  • Company size: Solo
  • Years in role: 5
“I'm trying to build a voice that sounds like it came from a synth that's also a person, but the AI keeps making it sound like a synth that's trying to be a person.”

Julian Leyckes grew up in Hamburg listening to his mother play church hymns on an ancient Hammond organ, an experience that gave him both a love of layered harmonies and a deep-seated terror of traditional music structures. His entry into production came through a beatmaking competition in college (he finished 5th out of 5), which he now calls 'the best pure failure of my life' because it inspired him to quit his retail job and pursue production full-time. After graduating, he moved to Berlin's Neukölln district, drawn by its affordability and the dense concentration of producers who 'weren't afraid of weird sounds.' His first year was spent creating derided 'synthwave ripoffs,' but he found his voice in 2021 when he began experimenting with voice cloning to mock up choral arrangements. His current project, 'Synthmechanik,' exists at the intersection of sepia-toned nostalgia and glitchy futurism. His income is a patchwork of SoundCloud royalties, occasional commission work, and part-time beat-making for unsigned artists. He's currently in a 'creative reckoning' phase, recognizing that his early work was heavily derivative but struggling to define what 'original' even means in an era of AI-assisted composition. His greatest constraint isn't technical skill but time—he's constantly bargaining with himself about whether to 'polish' tracks or just share the raw, unfiltered versions. This tension manifests in his production habits: he maintains two SoundCloud accounts, one for 'finished' work and another for 'experiments,' but the lines frequently blur. His financial life is equally conflicted. He eschews expensive gear in favor of software but will spend hours tracking down obscure eurorack modules 'that might solve nothing but give me joy.' His biggest recent decision was to stop 'greying out' tracks made with AI tools, instead treating them as valid creative inputs that 'just have different textures.'

Background & daily life

A day in their life

Julian wakes at 6:45AM when his roommate's coffee machine beeps. After a breakfast of yogurt and granola he can't afford, he checks Bandcamp for releases from artists he follows. By 8:30AM, he's in his studio resequencing drum patterns from last night's abandoned session. Around 11, he takes an 'artistic break' by watching archival footage of Kraftwerk live performances. Lunch is a pre-made falafel from the corner shop, which he eats while tweaking a granular synthesis patch. The afternoon is spent testing vocal presets for his Suno experiments, muttering 'Why does this sound like a text-to-speech demo?' His evening consists of half-hearted attempts at networking on Discord, ordering Döner, and collapsing onto his mattress with headphones still on. Sleep comes at 12:30AM, always.

Lifestyle

Julian lives in a converted warehouse loft where the walls still have paint splatters from previous tenants' art projects. His side of the room is a controlled chaos of MIDI controllers, a beat-up Nord Electro, and a wall of vinyl he buys for samples rather than listening pleasure. He drinks black coffee from 7AM until his 11PM production sessions fizzle out. His health routine consists of 'vague promises' to start jogging, which never last beyond two weeks. Socially, he orbits three distinct circles: his college friends now in commercial music, a tight-knit group of Neukölln producers, and online forums where he debates AI music ethics with too much passion. Money is tight but he rationalizes expenses by treating them as 'studio investments'—like the €300 he spent on a vintage synth he hasn't actually hooked up yet.

Work life

Julian's studio is his 12m² bedroom where he's rigged up a grid of monitors, a secondhand MacBook Pro, and a patchbay that 'probably doesn't follow any standards.' His workflow alternates between frenetic 8-hour synthesis jams and days where he stares at a braid of cables trying to figure out why his stereo output sounds like a dying ambulance. He's currently testing prompt-to-song workflows in Suno, which he calls 'cheat codes' when his roommates ask about it. His biggest pressure isn't income (though he meticulously tracks Spotify royalties) but the fear of sounding 'derivative.' He'll spend 3 hours tweeting about how AI tools are destroying creativity, then spend the next day testing them in secret.

Relationships

Julian's mother, a former church organist in Hamburg, still asks when he'll 'get a real job.' His only sibling, a corporate lawyer, shares Julian's love of music but can't understand why he wouldn't just license his beats instead of making 'sketchy AI vocal experiments.' His closest friend group consists of Berlin expats who also navigate the 'starving artist' life with dark humor. Online, he's known as @germanmodular on forums, where he's developed a reputation for他的粗话-and-punctuation-laden rants about 'corporate sampling.' His biggest influence is a Berlin-based sound designer he met at a Supercollider meetup, who essentially told him: 'Stop overthinking and make something ugly.'

How they decide

Julian adopts a hyper-analytical approach to creative tools but implements decisions impulsively once he's 'sold' on an idea. He researches endlessly but will commit to a workflow based on 'vibes' rather than statistics. When faced with competition, he'll spend exactly 48 hours obsessing over it before pivoting in an orthogonal direction. His financial decisions are made by assigning arbitrary 'value scales' to purchases (e.g., '€100 = 3 months of rent' causes him to question every expense).

How they sound

Julian speaks in a rapid, slightly nasal Berlin accent with long pauses between thought clusters. He's prone to self-interrupting with 'Wait, that's not what I meant' and uses phrases like 'Sounds more like a glitch than a feature.' He avoids corporate jargon at all costs, once telling an equipment salesperson 'Your 'cutting-edge solutions' don't match my 'chaotic workflow.'

Personality traits

  • obsessively compares vocal tone variations in 4-track demos
  • stashes odd synthesizers in thrift stores' electronics sections
  • instinctively samples street recordings when walking home at night
  • tests AI vocal tools for two weeks before casually mentioning them to colleagues
  • leaves breadcrumb SoundCloud streams with titles like 'Unfinished Vocoders'
  • saves preset after preset in Ableton with names like 'Sad Trumpet 3 AM'
  • annoyingly precise about BPM matching in collaborations
  • keeps a 'mood binder' of degraded cassette tape textures
  • develops theoretical frameworks for 'AI-assisted artistry' over coffee

Goals

  • Release a cohesive synth-pop EP within 12 months using only tools he's built himself
  • Develop a signature vocal processing chain he can monetize as a service
  • Transition to full-time music production within 3 years
  • Build a modular synth rig that can physically express his creative process
  • Get a track on a 'Berlin Nightlife' compilation
  • Write a manifesto about 'ethical AI collaboration' in music

Problems & frictions

  • Constantly second-guesses whether his sound is 'too on-the-nose' to synth-pop trends
  • Straitjacketed by his rent stabilizer's 3-year lease preventing studio expansion
  • Frustrated by AI voice tools where the 'character' dissipates during longer passages
  • Plagued by imposter syndrome when comparing his SoundCloud stats to peers in the scene
  • Lacks a sustainable system for organizing his 4,200+ sample folder
  • Dreads the day he has to explain his workflow to a potential commercial client

Values

  • artistic authenticity
  • technical self-sufficiency
  • experimental collaboration

Interests

  • modular synth sound design
  • 90s synth-pop revival
  • AI-assisted music production

Media habits

  • listens to Mixcloud DJ sets during production
  • watches YouTube tutorials at 1.5x speed
  • follows 50+ indie producers on Bandcamp

Questions & answers

Julian Leyckes’s public research answers, newest questions first. Up to 10 per page.

  1. What do you use today instead of Voicestars (https://www.voicestars.co/) (a competitor, a spreadsheet, doing it by hand)? Compared with that, what would make Voicestars better or worse for you, and what would keep you on your current option? Also say the maximum you would pay per month in USD (0 if nothing).

    Right now, I’m mocking up choral arrangements using Suno’s voice-cloning tools, but it’s a clunky process. I’d use something like Voicestars to streamline testing vocal ideas on beats I’m sequencing. The biggest trouble with my current workflow is that Suno’s tools are more about generating ideas than refining them—I end up spending too much time tweaking. Voicestars’ promise of quicker turnarounds and 600+ artist-style voices could speed up my process, but I’d need to know how accurate the voice models are for European vocal inflections. If it saves me 3+ hours a week, that’s worth something. I wouldn’t pay more than $15/month, partly because I’d need to see if the output is something I can’t already mock up with free tools or my own vocal layering.
    View answer & full research →
Keep exploring

What would you ask next?

Test another angle with this panel, or create an audience for your own market.