Build your LinkedIn presence on autopilot with posts that actually sound like you.

A blank chatbot prompt produces generic platitudes — it has no idea who you are, who you're writing for, or what already works in your niche. LinkedIn Persona starts with four things ChatGPT can't see: (1) your actual role and audience, (2) the structural patterns of high-performing posts in your specific vertical, (3) your real experience and opinions captured during onboarding, and (4) compliance guardrails for regulated industries. Every draft anchors in something specific about you instead of motivational filler.

They land closer to "you on a good writing day" than "AI-generated." Three things drive that: a two-pass writing model (Claude Sonnet draft → Claude Opus rewrite for natural cadence and sentence-length variance), a Persona Proof Layer that forces every post to anchor in something real (a client moment, an opinion you actually hold, a process detail), and a banned-phrase validator that flags engagement-bait and generic-AI tells before you ever see the draft. You review every post — and the system learns from your edits over time.

We use Perplexity's public-web search to find LinkedIn profiles and posts in your niche that are visible without logging in. We never scrape behind the LinkedIn login wall — that violates their TOS and we wouldn't survive a legal challenge if we did. From those public signals, we abstract the structural patterns (hook archetypes, length distributions, opinion shapes, CTA registers) into a playbook. The playbook never includes raw text from anyone else's posts — only the structural shapes. "Inspired by patterns, not copied from posts."

No — and this is deliberate, not a feature gap. The LinkedIn API requires Microsoft's explicit approval (which is gated on company size and revenue), and even with approval, auto-publishing tools tend to get flagged by the algorithm. So we generate drafts, you review and edit, you copy-paste into LinkedIn yourself. You stay in control of every post and your account never sees an API token.

No. We have no access to LinkedIn engagement data, your inbox, your private feed, or anything behind login. Zero. Everything we use is public information visible to anyone browsing without signing in. We'd need an OAuth integration to read any of that — and we don't have one.

Roughly 5 minutes. You'll need: your LinkedIn URL, a one-line description of your role, who you want to reach, and (optionally) a few profiles in your niche you admire. Step 3 runs an automatic profile enrichment that pre-fills the remaining steps so you're not staring at empty fields. You can finish in one session or come back later — your progress is saved.

Every LinkedIn post in those verticals runs through extra compliance validators before it reaches your review queue: no implied outcome guarantees ("we will cure", "guaranteed results"), no diagnostic claims ("if you feel X, you have Y"), no invented case studies, no testimonial-style claims without sourcing. Therapy posts also get flagged for shame-based or alarmist hooks. You set your vertical during onboarding and the checks fire automatically — no opt-in needed.

Free tier includes 1 LinkedIn generation per month — enough to see what a generated post in your voice looks like before you commit. Persona Pro (coming soon) unlocks unlimited generations, the full benchmark playbook, and weekly auto-generation on a schedule you set. No setup fees, no annual contracts, cancel anytime via the billing portal. We're a subscription tool, not an agency — you don't pay per-post.

Step 3 of onboarding runs an automatic discovery — we use Perplexity to surface LinkedIn profiles in your niche that are visible publicly. You pick which ones become your benchmarks; we never auto-include results. If you don't love any of the suggestions, you can skip the step entirely and the playbook will still work from your role + audience alone.

Yes — every draft lands in a review queue where you can edit text, hashtags, the alt-short version (for mobile preview), and the first-comment link before approving. If a draft misses the mark, you can reject it with a one-line reason and regenerate. The system learns from your edits: rejected drafts and edit patterns feed back into future generations so the next batch sounds more like you.

Yes. The Persona and local-business products share an account and the same admin workspace. You can run a personal-brand LinkedIn presence and a separate small-business profile from the same login — each is its own "Site" with its own settings.

Your onboarding data, generated posts, and edit history live in our Supabase database, encrypted in transit and at rest. We don't train shared models on your content — your voice profile is yours alone. Cancel your account and we delete your generation history on request. We don't sell data and we don't share it with anyone outside the platform.