Update README.md

README.md

@@ -1,11 +1,66 @@
 ---
 title: Api Transpetro Hackathon
-emoji: 🏃
+emoji: 🏆
 colorFrom: yellow
 colorTo: blue
 sdk: docker
 pinned: false
 license: mit
+short_description: API for predicting naval biofouling.
 ---
+Artificial Intelligence API for predicting naval biofouling, performance loss, 
+and financial risk in real-time. Developed based on naval physics, fuel chemistry, 
+and machine learning models (Random Forest + XGBoost).
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+The API is running in production on Hugging Face Spaces.
+
+Base Endpoint: https://carpenterbb-api-transpetro-hackathon.hf.space
+Interactive Documentation (Swagger): https://carpenterbb-api-transpetro-hackathon.hf.space/docs
+
+🔑 Authentication
+All requests must include the following
+
+access_token = hackathon_transpetro_2025
+
+🔮 Prediction Route
+POST /predict
+Sends data from a voyage or navigation segment and receives the hull diagnosis.
+
+{
+  "shipName": "Navio Teste 2",
+  "classe_navio": "Suezmax",
+  "speed": 12.5,
+  "duration": 24.0,
+  "distance": 300.0,
+  "beaufortScale": 3,
+  "Area_Molhada": 22500.0,
+  "MASSA_TOTAL_TON": 55.0,
+  "TIPO_COMBUSTIVEL_PRINCIPAL": "VLSFO",
+  "decLatitude": -23.0,
+  "decLongitude": -43.0,
+  "DiasDesdeUltimaLimpeza": 600
+}
+Response
+{
+  "status": "sucesso",
+  "navio": "Navio Teste 2",
+  "lof_previsto": 4,
+  "condicao_texto": "CRÍTICO",
+  "dados_tecnicos": {
+    "perda_performance_mj": 85000.50,
+    "risco_regional": 4.0,
+    "prejuizo_estimado_viagem_usd": 1340.25
+  },
+  "recomendacao": "AGENDAR LIMPEZA"
+}
+
+Author:
+Gabrielly Gomes
+Machine Learning Engineer.
+
+Attention:
+This model were developed for Transpetro(Petrobras) Hackathon.
+My team: 
+Gabrielly Gomes - ML Engineer
+Patrick Passos - Developer
+Lucas - Designer/Business