Data Engineering Platform

Cloud Retail Lakehouse

Designed a production-style lakehouse that ingests retail transactions, validates data quality, models dimensional marts, and serves executive reporting from a governed warehouse layer.

#AWS#dbt#Airflow#S3#Redshift

GitHub Live Demo

Overview

Designed a production-style lakehouse that ingests retail transactions, validates data quality, models dimensional marts, and serves executive reporting from a governed warehouse layer.

Problem

Retail leaders needed trusted daily KPIs across fragmented POS, inventory, and marketing exports without manual spreadsheet reconciliation.

Solution

Built automated ingestion, transformation, orchestration, and observability workflows that convert raw files into analytics-ready marts and Power BI dashboards.

Architecture

How the system is structured

S3 landing zone for raw transactional and reference data

Airflow DAGs orchestrating validation, dbt transformations, and warehouse loads

Redshift dimensional marts for sales, stock, and customer behavior

Power BI semantic model for margin, category, and regional performance

Tech Stack

AWSS3RedshiftAirflowdbtPower BI

Key Features

Incremental warehouse modelsData quality checks before publishExecutive KPI dashboardFailure alerting and lineage-friendly naming

Screenshots

Visual walkthrough

Revenue command center

Airflow orchestration graph

dbt model lineage

Power BI

Embedded dashboard

Public Power BI embeds can be connected here when a report URL is available.

Power BI

Cloud Retail Lakehouse Dashboard

Embedded report for the Cloud Retail Lakehouse case study.

Related Project