How We Build Vibe Cards for Public Figures

A personality vibe card is a compact, shareable read of who someone is based on how they actually talk — not a quiz they sat for, but the tweets, interviews, and long-form posts they have already published. When we build one for a public figure like Sam Altman or Naval Ravikant, the hard part isn’t making it look good. It’s making it defensible. This is the pipeline we run, and the lines we don’t cross.

What a personality vibe card actually is

Strip away the design and a vibe card is a synthesis: one page that triangulates several personality frameworks into a single coherent portrait, then backs each claim with evidence from the subject’s own words. For public figures we publish these in the Vibe Card Library — think of it as a personality database built from public language rather than fan speculation.

Every card scores the same four lenses at once, because no single framework captures a person. MBTI tells you how someone processes information; the Enneagram tells you what drives them; Big Five gives you measurable trait dimensions; and alignment captures their stance toward rules and people. Read together they triangulate. Read alone they mislead.

Step 1 — Sourcing public language only

For a private user, you upload your own messages. For a public figure, we are strict: public, on-the-record language only. No DMs, no leaks, no scraped private accounts. We gather a spread across formats and time so we’re reading a person, not a mood:

• Social posts — X, Reddit, and public captions, where someone writes fast and unfiltered.
• Long-form — essays, newsletters, and threads, where structured thinking shows.
• Transcripts — podcasts, interviews, and talks, where conversational rhythm comes through.
• Time spread — samples across months, so a single viral week doesn’t skew the read.

Why the spread matters: research on tweet-based personality assessment repeatedly flags that people express themselves differently by context — an introvert can read as extroverted in the right setting. Aggregating across formats and dates is the cheapest defense against that bias.

Step 2 — Reading the linguistic signals

Personality leaks through language in ways that are hard to fake. Decades of work using tools like Linguistic Inquiry and Word Count (LIWC) show that function words — pronouns, articles, hedges — carry more personality signal than the topics people choose. High first-person-singular use tracks with neuroticism; “we” tracks with agreeableness; negative-emotion words flag stress and instability. We read patterns, not single words:

Modern transformer models go further than dictionary look-ups, reading context and metaphor that keyword counters miss. We combine both — transparent heuristics for explainability, model scoring for nuance. If you want the full mechanics, the AI analysis breakdown and our piece on chat-based typing go deeper than we can here.

Step 3 — Scoring and cross-validation

Each lens gets scored independently, with a confidence level attached. Then the interesting work starts: making the four lenses agree, or admitting when they don’t. A directive, low-hedging communicator who frames everything in metrics should read as high conscientiousness on Big Five, a Thinking preference on MBTI, and likely a Type 3 or 8 color on the Enneagram. When the signals line up, confidence rises. When they conflict, we downgrade the claim rather than paper over it.

1. Score each lens. MBTI, Enneagram, Big Five / SLOAN, and alignment, each with its own confidence.
2. Cross-check. Do the lenses tell the same story? Reinforcing signals raise confidence; contradictions lower it.
3. Mark the soft spots. Thin or noisy areas get labeled as low-confidence instead of guessed.
4. Synthesize. Turn the numbers into a narrative plus practical scripts — how to work with this person, what to avoid.

The honesty here is the product. Observer-rated language studies explain a real chunk of personality variance — meaningfully more than self-report language does — but “a real chunk” is not “all of it.” A vibe card that pretends to certainty it doesn’t have is just astrology with better typography.

The limits we stay honest about

Building a personality vibe card responsibly means stating what it is not. A few hard rules:

We treat a public-figure card as a descriptive synthesis of a public persona — the version of themselves they chose to publish — not a verdict on the private human. For the deeper reasoning behind reading multiple frameworks together, see stacking typologies, and for the framework primers themselves, the MBTI guide, Big Five guide, and Enneagram guide.

A worked mini-example

Take a composite indie-hacker archetype — the kind of founder voice you see across X. The signals cluster fast: low hedging, directive tone, constant metrics framing, systems language, future-oriented planning. That profile reads as NT-flavored on MBTI, a 3-with-8 color on the Enneagram, and high Openness with low Neuroticism on Big Five. The resulting card doesn’t just label them — it ships a script: “Prefer async. Lead with constraints and deadlines. Keep meetings short.” That script is the part people actually use.

Browse real published examples in the Vibe Card Library, dig into the founder personality matrix, or check the broader field guide for more on how each framework holds up.