Embedding & vector search pipeline template

Components in this data flow diagram

• Docs, PDFs, product catalogue
• Parse & extract text
• Chunk ~500 tokens with overlap
• Attach metadata: source, date, ACL
• Embedding model: OpenAI text-embedding-3
• Upsert vectors
• Pinecone vector index
• BM25 keyword index
• User query
• Embed query
• ANN similarity search: top 50
• Keyword search
• Reciprocal rank fusion
• User allowed by ACL?
• Drop result
• Cross-encoder re-ranker
• Top 10 results
• Document updated or deleted
• Re-embed & update index

How to use this data flow diagram template

1. Click Open in the editor. The data flow diagram opens on a new page in FlowDiagram — nothing is uploaded and there is no sign-up.
2. Rename, add or delete shapes and drag them where you want; connectors stay attached as things move.
3. Drop in vendor logos from the library — 1,160 of them, including every AWS, Azure and Google Cloud service icon — or restyle it with the colour and theme controls.
4. Prefer text? Edit the Mermaid source below and paste it in through ☰ → Import diagram (see the Mermaid editor); it imports as an editable diagram.
5. Export to PNG, SVG or PDF, or press Ctrl+L for a link that carries the whole diagram.

Mermaid source

The whole template is the text below. Paste it into FlowDiagram, keep it in Git next to the code it describes, or hand it to an AI assistant to extend.

flowchart LR
  DOCS[Docs, PDFs, product catalogue] --> PARSE[Parse & extract text]
  PARSE --> CHUNK[Chunk ~500 tokens with overlap]
  CHUNK --> META[Attach metadata: source, date, ACL]
  META --> EMBED[Embedding model: OpenAI text-embedding-3]
  EMBED --> UPSERT[Upsert vectors]
  UPSERT --> VDB[(Pinecone vector index)]
  META --> KW[(BM25 keyword index)]
  QUERY([User query]) --> QEMBED[Embed query]
  QEMBED --> ANN[ANN similarity search: top 50]
  VDB --> ANN
  QUERY --> BM25[Keyword search]
  KW --> BM25
  ANN --> FUSE[Reciprocal rank fusion]
  BM25 --> FUSE
  FUSE --> FILTER{User allowed by ACL?}
  FILTER -->|No| DROP[Drop result]
  FILTER -->|Yes| RERANK[Cross-encoder re-ranker]
  RERANK --> RESULTS([Top 10 results])
  CHANGE[Document updated or deleted] --> REINDEX[Re-embed & update index]
  REINDEX --> VDB

Related: Data pipeline diagram tool · Mermaid editor · All ai & ml pipelines templates · All 521 templates