RAG Playground
Follow the pipeline from ingestion to cleaned text.
Use Backend
1) Ingestion
How data enters the system
1/10
2) Preprocessing & Cleaning
Needs file
Normalize and prepare extracted text
2/10
3) Chunking
Needs text
Split text into retrievable chunks
3/10
4) Embedding
Needs chunks
Convert chunks into vectors
4/10
5) Vector store / Indexing
Needs vectors
Persist vectors and payload mapping
5/10
6) Retrieval
Needs index
Search and retrieve relevant chunks
6/10
7) Re-ranking
Needs results
Reorder retrieved chunks for final context
7/10
8) Generation
Needs retrieval
LLM prompt and response generation
8/10
9) Evaluation
Needs generation
RAG metrics and observability
9/10
10) Experiment
Needs pipeline
Pipeline summary and analytics
10/10
Drag & drop your document
Upload PDF / DOCX / TXT to enter the pipeline.
Upload file
Load sample data