This section provides real-world examples of custom pipelines ranging from simple LLM wrappers to complex RAG architectures.
The best way to learn is by looking at the actual implementation in our repository:
NoOp
No retrieval, direct prompt to LLM.
View on GitHubarrow-up-right
Standard RAG
Document retrieval, vector store, and generation.
Deep Research
Advanced agentic workflow with research steps.
Tip: When developing your own pipeline, we recommend starting from the no_op folder as a template and progressively adding steps.
no_op
Last updated 28 days ago