# Open LakeHouse for East Africa Agricultural Market Data
A cloud-native, end-to-end agricultural **Lakehouse Data Platform** built on AWS that ingests, validates, transforms, and aggregates agricultural data across East Africa using a **Medallion Architecture (Bronze → Silver → Gold)**.
The project simulates real-world agricultural datasets including:
- Weather Data
- Crop Growth Data
- Farmer Profiles
- Field & Soil Data
The platform transforms raw agricultural datasets into **analytics-ready business datasets** for agricultural intelligence, climate-smart farming, and decision-making.
---
# Table of Contents
- Project Overview
- Architecture
- Key Features
- Tech Stack
- Project Structure
- Data Flow
- Bronze Layer
- Silver Layer
- Gold Layer
- Data Quality Validation
- AWS Services Used
- Step Function Orchestration
- Data Model
- Partition Strategy
- How to Run the Project
- Athena Queries
- Future Improvements
- Resume Highlights
- Author
---
# Project Overview
This project simulates a **production-grade agricultural data platform** for East African agricultural intelligence.
The platform:
1. Generates realistic agricultural datasets
2. Stores raw data in an AWS Bronze layer
3. Cleans and transforms data into a Silver layer
4. Performs automated data quality checks
5. Produces business-ready aggregations in a Gold layer
6. Makes datasets queryable using Amazon Athena
7. Sends failure notifications using SNS alerts
The architecture follows the **Lakehouse + Medallion Architecture Pattern** commonly used in modern enterprise data platforms.
---
# Architecture
## High-Level Architecture
```text
Synthetic Agricultural Data
│
▼
Python Data Generator
│
▼
Amazon S3 Bronze
(Raw JSON Data)
│
▼
AWS Glue Crawler
│
▼
AWS Glue ETL Jobs
Bronze → Silver
│
▼
AWS Lambda Checks
(Data Validation)
│
┌─────┴────────┐
│ │
Validation Pass Validation Fail
│ │
▼ ▼
Silver → Gold Amazon SNS Alert
Aggregation
│
▼
Amazon S3 G …