RAG pipeline & AI: Architect and develop the server-side RAG pipeline, covering query processing, embeddings and vector search through to prompt construction.
Full-stack development: Build both frontend and backend, ensuring that all LLM API calls are strictly handled server-side.
Subscriptions & payments: Implement a three-tier subscription model including a 7-day trial using Stripe.
Automation (Data Ingestion): Create a pipeline (e.g., using Make.com or n8n) that automatically transcribes new audio and video content (Whisper API), cleans it, splits it into chunks, and loads it into the vector database.
Data Intelligence: Develop an internal behavioral analytics dashboard that tracks search queries, emotional tone, content gaps, and usage-limit events in real time.
Requirements
Deep experience building modern full-stack web applications, ideally with our preferred tech stack: Next.js (React) for the frontend and Node.js (Express or Next.js API Routes) for the backend.
Practical experience building RAG systems, managing vector databases (Pinecone or Weaviate), and working with LLM APIs (notably Anthropic Claude Sonnet and OpenAI).
Proficiency with PostgreSQL (e.g., via Supabase) and authentication solutions such as Clerk or Supabase Auth.