eCommerce Microservices
May 23, 2025
·
1 min read

An e-commerce platform decomposed into independently deployable microservices. Each service lives in its own repository (wired in as Git submodules), owns its data store, and communicates through an API gateway and a message broker.
Architecture
| Service | Responsibility | Storage & messaging |
|---|---|---|
| Products | CRUD API for the catalog with business-logic and data-access layers (EF Core) and xUnit tests | MySQL, publishes events to RabbitMQ |
| Orders | Order creation and querying, RabbitMQ consumers, Ocelot API gateway for cross-service calls | MongoDB, Redis cache |
| Users | Authentication and user retrieval with a Dapper data layer | PostgreSQL |
| Frontend | Angular 17 single-page app that talks to the services through the gateway | — |
Deployment
docker compose -f docker-compose-build.yaml up --buildbuilds and starts the whole stack with seeded databases for local development.- Kubernetes manifests under
aks/run every service in Azure Kubernetes Service or any other cluster.
What it demonstrates
Service boundaries and polyglot persistence, asynchronous communication with RabbitMQ, API gateway patterns with Ocelot, containerization, and orchestrating a multi-service system with Docker Compose and Kubernetes.

Authors
Quantitative Analyst & Python/.NET Engineer
Quantitative analyst and .NET/Python engineer at Onward Energy in Denver. I build forecasting models,
market-data platforms, and the MLOps and CI/CD that keep them honest, backed by an M.S. in Data Analytics
(AI & ML, 4.0 GPA) and a decade of SQL Server, analytics, and full-stack work.