Nutrient Data Extraction API converts PDFs, scanned images, and Office files into spatial JSON or Markdown with bounding boxes, confidence scores, and page context. It supports four processing modes ranging from low-cost text extraction to AI-augmented parsing for complex layouts, handwriting, and formulas. Users can define a JSON Schema to map extracted values directly to their data model, enabling integration with databases, ERPs, CRMs, and AI pipelines. The API is designed for deterministic document workflows including RAG, search indexing, automation agents, and human review queues.