n8n Social Data Collection & RAG Pipelines
Two production n8n workflows built for a research client's internal use: one collects profiles and posts from several social platforms into MongoDB, the other turns that data into a retrieval-augmented chat agent over Qdrant.
Project Overview
Two production n8n workflows built for a research client's internal use: one collects profiles and posts from several social platforms into MongoDB, the other turns that data into a retrieval-augmented chat agent over Qdrant.
The Problem
Research and analytics work depended on fresh social data from several platforms, and analysts needed to ask questions of it in plain language. Collecting it by hand was slow and inconsistent, and a chat model without retrieval answered from guesswork.
The Solution
n8n as the orchestration layer for both ingestion and retrieval: scheduled or manual triggers, batched loops, code-node transforms, MongoDB as the store, and Qdrant with OpenAI embeddings behind an agent node.
Use Cases
Multi-platform collection
Separate branches per platform for profiles, posts, and keyword searches.
Batched, re-runnable loops
Items process in batches, and re-runs update records rather than duplicate them.
Normalised storage
Posts and profiles land in MongoDB in one consistent shape for analytics.
Vector index rebuild
Processed posts are split, embedded, and loaded into Qdrant in batches.
Grounded chat agent
Answers come from retrieved context, with short conversation memory.
Project Screenshots
What I Delivered
- n8n data-collection workflow across several social platforms
- MongoDB storage for normalised posts and profiles
- n8n workflow that builds a Qdrant index with OpenAI embeddings
- Chat-triggered RAG agent in n8n
Ready to build something useful?
Whether you need a freelance build, a full-time hire, or a short intro call, pick the path that fits and I will respond within 24 hours.