Organization card

README.md

@@ -1,10 +1,51 @@
 ---
 title: README
-emoji: 📚
-colorFrom: blue
-colorTo: red
+emoji: 📈
+colorFrom: indigo
+colorTo: green
 sdk: static
 pinned: false
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# QuantFeatures AI
+
+**Institutional-quality, point-in-time crypto market features from tick data, for independent quants and small teams.**
+
+Large funds have data teams that turn raw tick data into clean research features. QuantFeatures provides that feature data without a data-engineering team: trade flow, trade-size statistics, covariance, liquidations, funding, open interest, and top of book.
+
+## Free samples
+
+Hourly features for the first UTC day of each month, December 2019 to October 2026. Access is instant after a short form.
+
+| Dataset | Markets | Modules |
+| --- | --- | --- |
+| [binance-hourly-features](https://huggingface.co/datasets/QuantFeatures/binance-hourly-features) | Spot and perpetual (USDT) | 19 |
+| [okx-hourly-features](https://huggingface.co/datasets/QuantFeatures/okx-hourly-features) | Perpetual (USDT) | 17 |
+| [bybit-hourly-features](https://huggingface.co/datasets/QuantFeatures/bybit-hourly-features) | Perpetual (USDT) | 17 |
+| [hyperliquid-hourly-features](https://huggingface.co/datasets/QuantFeatures/hyperliquid-hourly-features) | Perpetual (USDC) | 15 |
+
+- **Point-in-time:** each row covers one instrument over a half-open UTC interval `[start, end)`; use `end_time_utc` as the availability time to avoid look-ahead bias.
+- **Explicit missing values:** NaN means not computable, never a silent zero.
+- **Easy to load:** Parquet, one config per module, one split per market.
+
+```python
+import pandas as pd
+df = pd.read_parquet("hf://datasets/QuantFeatures/hyperliquid-hourly-features/data/TradeFlow/perp.parquet")
+```
+
+Field definitions, timestamp semantics, and coverage: [Feature Catalog](https://github.com/QuantFeatures/FeatureCatalog).
+
+## Full data on request
+
+Continuous history for Hyperliquid and Binance, in two ways:
+
+1. **Delivered datasets:** we source the tick data and deliver features at intervals from 10 minutes to 3 hours.
+2. **Your own tick data:** if you license Tardis data, we compute features at any interval from 30 seconds to 1 hour.
+
+Real-time features run on your own infrastructure with your own data feeds.
+
+## Contact
+
+[QuantFeatures@gmail.com](mailto:QuantFeatures@gmail.com) · [GitHub](https://github.com/QuantFeatures)
+
+*QuantFeatures is a trade name of ValueFuse LLC. Research data, not trading recommendations or a promise of predictive performance.*