Update README.md

README.md

@@ -1,3 +1,50 @@
----
-license: apache-2.0
----
+---
+license: apache-2.0
+task_categories:
+- text-classification
+language:
+- en
+tags:
+- techcrunch
+- news
+- startup
+- founders
+- pitch
+- pitch videos
+pretty_name: TechCrunch Battlefield News
+size_categories:
+- 1K<n<10K
+---
+
+# TechCrunch Battlefield News — Preview
+
+This repository hosts a small preview of a structured dataset that tracks companies featured in TechCrunch’s Startup Battlefield coverage. The preview is here so you can test the schema and sample rows locally. To purchase the complete dataset with full coverage and updates, go to **https://www.thedataoutlet.com**.
+
+> **Buy the full dataset:** https://www.thedataoutlet.com
+
+---
+
+## What you get in this preview
+
+- A single CSV file with 62 rows and 19 columns.
+- Fields include company and article metadata like title, author, event name, category, location, money raised, links, and a collected timestamp.
+- Values are not cleaned for every edge case in the preview. Expect occasional HTML entities or nulls. The paid dataset ships with tighter normalization.
+
+**File list**
+
+- `TechCrunch_Battlefield_News.csv` 62 rows
+
+## Quickstart
+
+Load directly with `datasets`:
+
+```python
+from datasets import load_dataset
+
+ds = load_dataset(
+    "csv",
+    data_files="TechCrunch_Battlefield_News.csv",
+    split="train"
+)
+
+print(ds[0])