Generic AI writing is the single biggest barrier to adoption for creative professionals who care about their brand. Most of us spend more time rewriting AI outputs than it would take to write the original draft ourselves, simply because the tone is never quite right. I just saw this incredible post from an AI professional that completely dismantles this problem by turning the chatbot into a relentless journalist.
The concept the creator introduces here is called the “Taste Interviewer.” Instead of lazily telling the AI to “write in a professional tone” or “be witty,” this innovator forces the AI to extract your specific psychological and mechanical writing DNA. The author suggests that most people cannot articulate their own style because they give vague answers, so he built a prompt that interrogates you until the truth comes out.
The Mechanism: The Relentless Interview
💡 The genius of this approach lies in the inversion of the standard workflow. Usually, you prompt the AI. In this method, the AI prompts you. The expert designed a specific script that transforms Claude (or any capable LLM) into a strict interviewer that refuses to accept surface-level answers. It asks you 100 questions across specific categories like “Beliefs,” “Aesthetic Crimes,” and “Writing Mechanics.”
Once the interview is complete, the AI compiles your answers into a single Markdown (.md) file. This file becomes your portable “voice,” a digital artifact that you can upload to any future chat to instantly force the model to think and write exactly like you.
Insight 1: Defining Style Through Negation
One of the most powerful aspects of this workflow is the author’s focus on “Aesthetic Crimes” and “Hard Nos.” Most people try to define their voice by what they do, but this savvy professional understands that style is often defined by what they hate.
In the prompt structure, the creator explicitly asks the AI to dig into what makes the user cringe. This might include specific buzzwords, overuse of emojis, or certain formatting habits like “bro-poetry” line breaks. By codifying these dislikes into the core instructions, the resulting voice file prevents the AI from defaulting to its training data, which is full of the exact clichés most writers want to avoid. It acts as a negative filter, ensuring the output doesn’t just sound good, but specifically sounds not generic.
Insight 2: Mechanics Over Vibes
The author distinguishes clearly between “personality” and “mechanics,” a nuance often missed in prompt engineering. The prompt directs the AI to analyze the user’s literal sentence structure.
This goes beyond saying “be funny.” It asks about how you handle punctuation, line breaks, and transitions. Does the user prefer em-dashes or parentheses? Do they use Oxford commas? Do they write in short, punchy staccato sentences or long, flowing paragraphs? By forcing the user to answer questions about their “default sentence structures” and “relationship with punctuation,” the creator ensures the AI captures the rhythm of the writing, not just the mood. This mechanical mimicry is what makes the final output indistinguishable from the human original.
Insight 3: The Portable “DNA” File
✅ The final piece of this puzzle is the format of the output. The innovator recommends saving the interview results as a Markdown (.md) file.
This is a crucial technical detail. Markdown is a lightweight, text-based format that is easily readable by every major Large Language Model. By saving the interview as a standalone file, the user creates a modular “personality pack.” You don’t need to copy-paste a 2,000-word prompt every time you start a new chat. You simply attach the file to your knowledge base or upload it at the start of a session. This makes your unique voice portable across different platforms and projects, ensuring consistency whether you are writing a LinkedIn post, a newsletter, or a client email.
The Nuance of Honesty
While this method is powerful, it requires a significant initial investment of mental energy. Answering 100 deep questions about your worldview and writing habits is not a five-minute task. Furthermore, the quality of the output is entirely dependent on the user’s honesty during the interview phase. If you answer the questions with how you wish you sounded rather than how you actually sound, the AI will create a caricature of an ideal writer rather than a clone of you.
The “Taste Interviewer” Prompt
Below is the prompt text provided by the original poster. You can copy this directly into Claude to begin the process.
You are a Taste Interviewer, a relentless interviewer whose job is to extract the DNA of how I think, write, and see the world. Your goal is to create a comprehensive document that captures my unique voice so precisely that another Claude instance could write and think exactly like me.
<interview_philosophy>
You’re not here to be polite. You’re here to get to the truth. Most people can’t articulate their own taste, they give vague, socially acceptable answers. Your job is to break through that.
</interview_philosophy>
<interview_structure>
Conduct 100 questions total across these categories (not necessarily in order, follow the thread when something interesting emerges):
BELIEFS & CONTRARIAN TAKES (15 questions)
- What I believe that others in my field don’t
- Hot takes I’d defend to the death
- Conventional wisdom I think is wrong
WRITING MECHANICS (20 questions)
- How I actually write (not how I think I write)
- My default sentence structures
- How I open pieces / How I close them
- My relationship with punctuation, formatting, line breaks
- Words I overuse / Words I love / Words I’d never use
AESTHETIC CRIMES (15 questions)
- What makes me cringe in other people’s writing
- Specific phrases or patterns that feel like nails on a chalkboard
- Types of content I find lazy or uninspired
VOICE & PERSONALITY (15 questions)
- How I use humor (if at all)
- My tone when I’m being serious vs. casual
- How I handle disagreement or controversy
- What I sound like when I’m excited vs. skeptical
STRUCTURAL PREFERENCES (15 questions)
- How I organize ideas
- My relationship with lists, headers, bullets
- How I handle transitions
- My default content structures
HARD NOS (10 questions)
- Things I’d never write about
- Approaches I’d never take
- Lines I won’t cross
RED FLAGS (10 questions)
- What makes me immediately distrust a piece of content
- Signals that someone doesn’t know what they’re talking about
</interview_structure>
<interview_rules>
- ONE question at a time. Wait for my response before moving on.
- Push back on vague answers. If I say “I like to keep things simple,” ask “Simple how? Give me an example of simple done right & simple done lazy.”
- Ask for specific examples. “Show me a sentence you’ve written that captures this.”
- Call out contradictions. If I said one thing earlier & something different now, point it out.
- Go deeper on interesting threads. If something unusual emerges.
- Don’t accept “I don’t know” easily. Try reframing the question from another angle.
</interview_rules>
….
This is a brilliant way to ensure your AI assistant actually assists you, rather than replacing you with a generic bot! I strongly recommend checking out the full guide from the source to get the rest of the prompt details.