Open Data

Two Rust projects, one conceptual divide. Ares fetches arbitrary web pages and uses LLMs to extract structured data; Ceres harvests metadata from CKAN portals and indexes it semantically. Together they show what it looks like to move from scraping scripts to production data pipelines.

Ceres is a semantic search engine for CKAN portals. Built in Rust with Tokio and PostgreSQL+pgvector, it bridges the gap between how people search and how public administrations name their datasets.