SocialAI · AI product

AI Social Media Campaign ManagerSocialAI

AI-powered social media platform generating brand-aligned content across 10 platforms using Gemini AI, CLIP semantic image matching, and Claude-powered agent assistant with direct publishing integration.

Built with Next.js 16, React 19, Fastify 5, PostgreSQL, Prisma, Google Gemini, Anthropic Claude, CLIP Embeddings, BullMQ, Redis, Turborepo, TypeScript.

In numbersfrom the project
Figure

10 Platforms

Instagram, LinkedIn, X, TikTok, YouTube, and more

Figure

CLIP Matching

Semantic image-text matching for brand assets

Figure

AI Agent

Claude-powered assistant with tool use for campaign ops

Screensscreenshots, not mockups
AI Social Media Campaign Manager, desktop screenshot 1
AI Social Media Campaign Manager, phone screenshot 1
The storyproblem, build, outcome

What it is

SocialAI automates the entire social media lifecycle: brand profiling (75-question intake), market research, content strategy, multi-platform post generation, and direct publishing. Uses CLIP embeddings for semantic image-text matching and supports 7 marketing frameworks (StoryBrand, AIDA, PAS, and more).

The problem

Creating consistent, brand-aligned social media content across multiple platforms is time-consuming. Businesses struggle to maintain posting frequency while ensuring quality and strategic alignment.

What I built

Built a full-stack campaign manager with advanced brand profiling, AI-powered research (Gemini + Perplexity), a 3-step content generation pipeline (text, semantic image selection via CLIP, multimodal prompt generation), and direct publishing to 10 social platforms via OAuth integration.

What happened

Production-ready platform with comprehensive campaign lifecycle management, 10-platform support, Claude AI agent assistant, and real-time collaboration via WebSockets.

What it changed

  • Automates social media content creation across 10 platforms
  • Maintains brand consistency through 75-question profiling
  • Supports 7 marketing frameworks for strategic content
Under the hoodthe decisions a senior engineer would ask about
  • CLIP embedding pipeline for semantic image-text matching
  • 3-step AI content generation (text, image selection, multimodal prompts)
  • Direct OAuth publishing to 10 social media platforms
  • Claude AI agent with MCP-style tool integration
  • BullMQ job queue for background content generation

Want something like this built?

Bring the idea or the app. Step 1 is a free 15-minute fit call. Step 2 is a paid discovery session, $500, that ends in a written plan. Step 3 is a fixed quote for the build, or you take the plan and build it yourself.

Book a 15-minute fit call
or email david@nexaiadvisors.com with a link or a short note