Skip to main content

Web2026Live product

Cinelas — Movie & TV Discovery

A live film and TV platform for discovering titles, tracking what you watch, and finding where to stream them.

Platforms
Web
Focus
Search in your own words · 1.3M+ titles · From discovery to watchlist

Project view

Cinelas home page showing rows of movie and television recommendationsVisit the live project (opens in a new tab)
The live discovery experience at cinelas.com.

The story

The problem

Finding something to watch often means switching between streaming services, recommendation sites, and personal watchlists.

What it delivers

I built Cinelas to bring discovery, title details, personal lists, viewing progress, and multi-category reviews into one responsive application.

At a glance

  • Search in your own words

    Ask for a film by mood, story, actor, or a scene you remember instead of relying only on filters.

  • 1.3M+ titles

    Search a catalogue of more than 1.3 million films and series.

  • From discovery to watchlist

    Find a title, check where it streams, save it to a list, and track or review it later.

Built with

  • RAG
  • .NET
  • React
  • Azure
  • Docker
  • pgvector
  • TypeScript
  • PostgreSQL
  • Tailwind CSS
  • Azure OpenAI
  • Grounded recommendations (RAG)

    A retrieval-augmented generation (RAG) workflow searches a prepared index of film and TV records for relevant matches. Azure OpenAI uses the retrieved titles to select and explain recommendations, keeping suggestions grounded in catalogue data.

  • Checks around AI output

    The recommendation flow filters out watched titles and verifies suggestions against the catalogue before displaying them. Authentication, usage limits, caching, and monitoring help manage performance and API costs.

A closer look

Uses catalogue matches to support recommendations from natural-language requests.
A review captures more nuance than a single star score.

Continue exploring

More projects

View all projects →︎