← All selected work
02 / Backend engineering

EventHub

A Python API for events and RSVPs, with authentication, resource ownership and traceable imports from Leeds City Council.

PythonFastAPIPostgreSQLSQLAlchemyPytest
Academic project
Context
Academic project
My contribution
Built the API, access controls, import workflow and automated checks.
Evidence
24 REST endpoints
THE PROBLEM

Start with the need.

An event API needs more than basic create, read, update and delete operations. It needs to handle who can change each resource, import external records safely and explain the provenance of those records.

My contribution

I built an event-management backend for University of Leeds coursework, covering events, attendees, RSVPs, analytics and administrative imports. I integrated Leeds City Council event data and added provenance records, authentication, access controls and automated checks.

How it works

STEP 01Client request
STEP 02Auth & ownership
STEP 03FastAPI services
STEP 04PostgreSQL

Conceptual flow. Implementation details and reproducibility instructions are in the repository.

A closer look at the work.

EVENTHUB / DOCUMENTED API SURFACEThree of 24 routes
GET/eventsExplore the event collection
POST/events/{id}/rsvpsCreate a reservation
GET/events/{id}/provenanceInspect an imported event’s origin

Routes are taken from the repository documentation. Authentication and ownership requirements vary by operation.

The decisions behind the code.

Check ownership as well as roles

Resource-level owner checks complement JWT authentication and administrator permissions.

Keep imported data explainable

Provenance records connect imported events to their source dataset and import batch.

Support development and deployment

Dialect-aware queries support SQLite locally and PostgreSQL, with Alembic migrations and CI checks.

What the evidence says.

The public README documents 24 REST endpoints, 84 automated tests and approximately 95% test coverage. It also describes Ruff, mypy and Bandit checks. These are repository-reported figures; coverage is not a guarantee of complete correctness.

Public repository README, reviewed 17 November 2026.

Read the source documentation

Where the boundaries are.

  • This is an academic backend project, not a commercial event platform with established customers.
  • The historical Render deployment has not been verified for this portfolio, so the repository is the primary link.
  • Test coverage measures execution of code during a test suite, not every possible failure or security issue.

What I learned

Integration work becomes easier to explain and debug when ownership, provenance and error behaviour are designed into the API.

UP NEXT

Ledger Data Quality

Read the next case study
OPEN TO ROLES & PROJECT CONVERSATIONS

Have a role or
project in mind?

Get in touch