Sportopod

Sportopod manual

A short guide to the archive: what it is, how to find your way around it, and what the demo showed.

Updated

In this manual

What the archive is

Sportopod is a sports-news site we built to show our LLM layer and smart router at work. It gathered stories from many outlets, grouped them by event and published one checked article per event, refreshed every six hours while it ran.

New stories have not been collected since July 2026. The site stays up as a frozen demo archive: everything you can read there is what came out of the pipeline.

About the pictures

The screens on this page were captured on 22 September 2026, with each article’s photo. The site today shows its logo in place of most photos.

Browse it

Everything is one or two clicks from the home page. No account, nothing to install.

  1. Open sportopod.com. The home page leads with the top story and the trending ones.

    Sportopod home page: the lead story and the trending stories, each with its photo
  2. Pick a sport: Football, Basketball, Soccer, Baseball, Hockey, Motorsport, Combat, Tennis, Golf, Cricket, Rugby or the Olympics, and inside them F1, NASCAR, IndyCar, MotoGP, MMA and Boxing.

  3. Or pick a kind of story: trade rumors, injuries, recaps, previews, rankings, opinion, deals and interviews. Scores have their own page; the standings page is in the menu but has no tables in the archive.

  4. The HTML site map lists every section on one page.

Read an article

Every article was built the same way from its sources.

  1. The story comes first: one article for one event, written from the outlets that covered it, often in several languages.

    The top of a Sportopod article: the headline, the photo with its credit, and the start of the story
  2. Under it: a short “why this matters”, the questions a reader would ask, and the original sources, linked.

    Further down a Sportopod article: why it matters, the questions readers ask, and the source

What the demo showed

Sportopod was a hard, public test for the AI plumbing we build for clients. News is fast, comes in several languages, and a wrong article is visible to everyone.

One event, one cluster

Stories about the same event were grouped together, so the reader got one article, not twenty.

The router

Our router chose the model for each task by itself.

The schema check

Every answer from a model had to arrive in an agreed structure: headline, summary, body, source and so on. An answer that did not fit was not published.

For search engines and AI assistants

A sitemap, a news sitemap, an RSS feed of the newest articles and an llms.txt, a plain map of the site for AI assistants.

Under it

The pipeline runs on PostgreSQL with pg_cron and pgvector. The models are reached through our own LLM gateway, written in TypeScript on Fastify 5. Six branded sites ran from one codebase.

Privacy

Reading the archive needs no account. Sportopod’s own privacy policy and terms are on the site.

Contact

A question about the archive, or a pipeline like this for your own content or data: write to us.

ByFounder and Imaginator

Pineido, based in Strumica, North Macedonia, is a software company that builds websites, online shops, web and mobile apps, business automation and AI for companies at home and abroad. Work ships in 30-day sprints at a fixed price, and the client owns the code.

Published Updated