Uploaded March 2026 | Updated September 2026, 2 weeks ago

---

Learn how to use Extractors in H2O Enterprise h2oGPTe to automatically convert unstructured documents into clean, structured JSON data — no manual reading required.

In this tutorial, we extract key financial ratios from Alphabet's Form 10-K using a JSON schema and an LLM, walking through every step from creating a Collection to viewing your structured output.

What you'll learn:
→ What Extractors are and how they work
→ How to create a Collection and upload a document
→ How to define a JSON schema for your target fields
→ How to run an extraction and view structured JSON results

Ideal for: data engineers, analysts, and developers working with business documents like SEC filings, invoices, or resumes.

──────────────────────────────
🔗 Resources & Links

H2O Enterprise h2oGPTe: h2o.ai/platform/enterprise-h2ogpte
H2O.ai Documentation: docs.h2o.ai/enterprise-h2ogpte
Download the sample Form 10-K used in this tutorial: s206.q4cdn.com/479360582/files/doc_financials/2025/q4/GOOG-10-K-2025.pdf
H2O.ai Platform Overview: h2o.ai

──────────────────────────────
📌 Chapters

00:00 Introduction
00:15 What are Extractors?
00:30 Step 1 — Understand the objective
00:50 Step 2 — Create a Collection
01:30 Step 3 — Create your Extractor
02:30 Step 4 — View extracted results
03:00 Closing

──────────────────────────────
H2O.ai builds AI platforms that help enterprises move faster with their data. Enterprise h2oGPTe is H2O.ai's secure, document-aware AI platform for enterprise teams.

#H2OAI #h2oGPTe #DataExtraction #LLM #EnterpriseAI #JSON #DocumentAI