Metadata Enrichment Pipeline
Automatically enrich your data with extracted metadata: entities, topics, sentiment, language, and custom attributes. Transform raw content into structured, queryable data.
from mixpeek import Mixpeekclient = Mixpeek(api_key="YOUR_API_KEY", namespace="enriched-data")# 1. A bucket for feedback, and a collection whose LLM step extracts entities,# topic, sentiment and language onto every documentbucket = client.buckets.create(bucket_name="customer-feedback",bucket_schema={"properties": {"ticket": {"type": "text"}}},)collection = client.collections.create(collection_name="customer-feedback",source={"type": "bucket", "bucket_ids": [bucket["bucket_id"]]},feature_extractor={"feature_extractor_name": "text_extractor","version": "v1","parameters": {"response_shape": "entities (list of names), topic, sentiment (positive, neutral or negative) and language code","llm_provider": "openai","llm_model": "gpt-4o-mini",},},)# 2. Upload and processclient.buckets.upload(bucket["bucket_id"],blobs=[{"property": "ticket", "type": "text", "data": "s3://your-bucket/feedback/ticket-8812.txt"}],)client.collections.trigger(collection["collection_id"])# 3. Read the enriched documentsdocs = client.documents.list(collection["collection_id"], page_size=100)for doc in docs["results"]:print(doc["document_id"], doc.get("sentiment"), doc.get("topic"), doc.get("entities"))
Feature Extractors
Text Embedding
Extract semantic embeddings from documents, transcripts and text content
Retriever Stages
Frequently asked questions
What entities are extracted?
People, organizations, locations, dates, monetary values, products, and custom entity types.
Can I combine multiple enrichment types?
Yes. Apply as many extractors as needed. Each runs independently and results are merged into the document metadata.
Does enrichment work in real-time?
Yes for text content. Image and video enrichment is near-real-time depending on the extractors used.
Use Cases Using This Recipe
Contextual Page Signals for Ad Placement
One article URL in, five structured signals out: IAB category, entities with salience scores, sentiment, brand safety, keywords.
Five, from a single pass
Signals per URL
Contextual advertising and ad-targeting teams who decide, per article, which creative renders beside it, and who need that decision to be defensible to a brand.
Related Recipes & Resources
Explore these related resources to deepen your understanding and discover more powerful features
Text Embedding
Extract semantic embeddings from documents, transcripts and text content
Multimodal RAG Pipeline
Build a retrieval-augmented generation system that works with text, images, and video. Feed relevant multimodal context to LLMs for grounded responses.
Taxonomy Enrichment Pipeline
Automatically classify and tag content using custom taxonomies. Map your content to IAB categories, custom hierarchies, or industry-specific classifications.
Content Clustering Pipeline
Automatically group similar content together using embedding-based clustering. Discover themes, identify duplicates, and organize large content libraries.
Multimodal Knowledge Base
Consolidate documents, videos, images, and audio into a single searchable knowledge base with RAG capabilities. Supports natural language Q&A across all content types, with citations linking back to the exact source document, video timestamp, or image.
Feature Extraction
Multi-tier feature extraction that decomposes content into searchable components: embeddings, transcripts, detected objects, OCR text, scene boundaries, and more. The foundation for all downstream retrieval and analysis.