Become a social media influencer without the daily grind. RapidContent does the grind work — finding angles from recent news in your niche, drafting posts in your voice, formatting, and producing media — while the main voice and ideas stay yours. You review and approve; everything else is automated.
The workflow is built around minimal decisions: the AI proposes topic ideas, you pick one, it writes the full post, you approve, and it's saved to your content library ready to publish.
The AI-social-content space is crowded, so RapidContent is built toward a wedge, not a generic clone:
- Privacy-first / on-prem capable. The LLM layer is vendor-agnostic and supports local models (Ollama), so the engine can run fully on-prem with no external calls — for enterprises, agencies, and founders who can't send data to cloud SaaS.
- News + your opinion, not generic summaries. Ideas come from recent news in your niche, run through your profile (expertise, point of view, voice) to produce takes that sound like you.
- Built toward true end-to-end, multi-platform. Starts with LinkedIn text posts, architected so platforms (X, Threads, Instagram, blogs…) and formats (text, carousel, video) are pluggable.
- Next.js (App Router) + React, Tailwind — deployed on Vercel
- Postgres + Prisma (Neon)
- Auth.js (NextAuth) + Google OAuth
- Vercel AI SDK behind a swappable multi-provider LLM interface (Gemini default; Claude, OpenAI, DeepSeek, Ollama supportable)
- News via curated RSS feeds (on-prem) with an optional API fallback
npm install
npm run devOpen http://localhost:3000. See .env.example for the
environment variables you'll need (database, OAuth, LLM, and news keys).
Contributions are welcome! Whether it's a new LLM provider adapter, support for another platform, a bug fix, or docs — feel free to jump in. Open an issue to discuss an idea, or send a pull request. If you're not sure where to start, open an issue and let's talk.
Najam — najam.pk
An agentic AI developer and architect focused on enterprise and on-prem AI.