Skip to main content
Core Studio Architecture

Engineered for authentic social publishing.

Explore the 5 core AI and rendering subsystems that turn raw thoughts into voice-matched posts and designed PDF carousels.

Studio Capabilities

5 engines built for speed and authenticity.

Every layer of PrimusPost is designed to eliminate manual formatting while preserving your authentic writing style.

Voice Engine · Subsystem 01

Voice DNA Pattern Calibration

Analyzes your sentence length patterns, vocabulary choices, paragraph pacing, and hook structures from past writing. Generates new posts in your authentic voice rather than generic AI.

  • Sentence length, tone, and vocabulary analysis
  • Multiple voice profiles for different brands or clients
  • Learns from edits to refine voice accuracy over time
  • Enterprise AI endpoints with zero model training
Multi-Channel · Subsystem 02

1 Input ➔ 5 Native Formats

Generates tailored post structures for LinkedIn, X (Twitter), Instagram Captions, Meta Threads, and designed PDF carousels from one paste.

  • LinkedIn posts with structured line breaks and hooks
  • X standalone posts optimized for 280 characters
  • Threads conversational hook structuring
  • Instagram captions with clean hashtag trees
Design Studio · Subsystem 03

Typography-First PDF Carousels

Generates PDF carousels directly from text using a programmatic Satori layout engine. Includes embedded selectable text layers ready for LinkedIn document upload.

  • 8 curated design themes: Modern, Professional, Bold, Split, Ribbon, Glass, Elegant, Tech
  • 14 slide layouts: covers, numbered steps, quotes, stat callouts, checklists, comparisons, before/after, charts and more
  • Automatic line balancing and font scale adjustments
  • PDF export with a selectable, searchable text layer
Quality Guardrail · Subsystem 04

Anti-Slop AI Defense

An automated filter that removes overused AI buzzwords, corporate filler, and cliché phrases before content is presented for review.

  • Active filter for overused words (delve, tapestry, game-changer, unlock)
  • Removes artificial conversational filler and LinkedIn clichés
  • Preserves direct, human sentence structures
  • Grounds output strictly in your input facts
Topic Intelligence · Subsystem 05

Pulse Signal & Idea Matching

Surfaces relevant topic signals and angles in your niche to spark ideas for your next content batch.

  • Browse niche topic signals and talking points
  • Generate fresh angles on demand from any signal
  • Filter by industry and content theme
  • Save draft angles directly to your content library
Execution Pipeline

How raw ideas become publish-ready content.

The sequential processing pipeline from input ingestion to final PDF compilation.

Stage 01

Ingestion & Parsing

Accepts raw text drafts, voice transcripts, or bullet notes and parses core arguments.

Structured Source Draft
Stage 02

Voice & Anti-Slop Calibration

Applies your saved Voice DNA profile while actively stripping overused AI clichés.

Voice-Matched Copy
Stage 03

Multi-Channel Synthesis

Formats copy into native structures for LinkedIn, X, Threads, and Instagram.

4 Social Post Formats
Stage 04

Satori Slide Compile

Renders typography-balanced slides with embedded text layers for direct document upload.

LinkedIn-ready PDF
Technical Specifications

Platform specifications and rendering standards.

Supported formats, models, resolution standards, and privacy terms.

DimensionSpecification & Architecture
Supported Input TypesText notes, bullet lists, voice transcripts, meeting takeaways, article drafts
Generated Social FormatsLinkedIn Posts, X Posts, Instagram Captions, Threads Posts, PDF Carousels
AI Processing ModelGoogle Gemini 3.6 Flash via Vertex AI (Enterprise Tier, No Model Training)
Carousel Rendering EngineSatori + resvg-js + @cantoo/pdf-lib (PDF with a selectable, searchable text layer)
Carousel Aspect Ratio1080 × 1350 px (4:5 Aspect Ratio — LinkedIn & Instagram Standard)
GenerationAll 5 native formats produced in parallel from a single run
Privacy & Data ProtectionEnterprise API terms, encrypted in transit and at rest, full user ownership
Technical FAQ

Frequently asked feature questions.

How is PrimusPost different from standard ChatGPT or Claude prompts?

Standard chat models require manual prompting and produce generic, recognizable AI phrasing. PrimusPost extracts your personal Voice DNA from writing samples and applies automated anti-slop filters while generating 5 platform-native formats and PDF carousels in one pass.

Do I need design tools like Canva to make carousels?

No. PrimusPost uses a programmatic typography engine (Satori) with 8 design themes that automatically formats your slides with balanced line heights, contrast, and font scales.

Is the text in an exported carousel selectable?

Yes. Every carousel exports as a PDF with a selectable, searchable text layer, designed specifically for LinkedIn document posts.

Are my unpublished drafts used to train AI models?

No. We operate on enterprise-tier paid AI APIs where user drafts, proprietary ideas, and voice profiles are never used to train public foundation models.

Your next post should sound like you.

Turn one rough idea into 5 native formats from a single paste. 15 free generations to start, then 3 every month. No credit card required.