Description:
We are seeking an experienced AI developer to set up a Retrieval-Augmented Generation (RAG) pipeline for a small video agency. The goal is to integrate our 30 blogs and brand voice document into a system that can suggest recommendations for tone, style, and wording when creating future content.
The project will require:
• Embedding the blogs and brand voice/identity document using OpenAI or similar tools.
• Storing the embeddings in a vector database (e.g., Pinecone, Weaviate, or FAISS).
• Creating a simple RAG workflow to retrieve and utilize stored content during text generation.
• Optionally integrating with LangFlow for a user-friendly interface.
The ideal candidate has:
• Experience with LangChain, LlamaIndex, or Haystack.
• Familiarity with vector databases like Pinecone or FAISS.
• Hands-on experience with LLM APIs (OpenAI, Anthropic, etc.).
Deliverables:
• Fully functioning RAG pipeline with clear instructions for use.
• Support for 30 blogs and one brand document.
This is a one-time project with the potential for ongoing support if needed.
Reasonable Hours
• 10–15 hours for a skilled developer to set up and test a basic RAG system.
• Add 2–3 hours for extra features or customization (e.g., LangFlow integration).
• Please provide examples of similar work.
• Test Run: Request a short demo or outline before committing to the full project.
