Flight Operations System

A transactional flight-booking system built for NYU's Database Systems final — a real relational schema, staff and customer roles, and parameterized queries under concurrent writes.

Source

A course final built around the part database classes actually test: can you design a normalized relational schema and then write correct, parameterized queries against it under real use cases — search flights, book a ticket, rate a flight, report on staff performance.

The schema (schema.sql) and seed data (seed.sql) are both declarative and checked into the repo, alongside a CSV import script for loading flight data and a written set of use cases and queries documenting what each endpoint has to support. Staff and customer roles get separate authenticated flows rather than one shared view with conditionals sprinkled through it.

Finished and graded, not maintained since — left up because the schema design and the query set are the actual deliverable, more than the templated Flask views around them.

What it does

Nine-table relational schema (schema.sql) with declarative seed fixtures (seed.sql)
Parameterized queries throughout — no raw string interpolation into SQL
Separate staff and customer authenticated flows
Flight search, booking, and ratings
CSV import script for loading flight data

What it taught me

Normalizing a relational schema for a domain with real many-to-many relationships
Writing parameterized queries by hand instead of leaning on an ORM
Translating written use cases directly into the queries that satisfy them