Changelog
Product updates and improvements
A running log of what changed in the dataset, schema, and enrichment models. Entries are tagged so you can scan for what matters to your work:
- new: additions to the dataset, new fields, new products, or new endpoints.
- improved: accuracy gains, coverage expansion, or model retraining.
- fixed: corrections to historical data, schema bugs, or enrichment edge cases.
Subscribers receive the same notes via email when changes ship. For accuracy benchmarks behind each model release, see data quality; for the field-level schema, see schema.
Title-Aware Skills Relevance Filter
Shipped a title-aware relevance filter for skills extraction. After the high-recall dictionary scan tags candidate skills, the filter reviews each role and skill pairing and removes off-role mentions (e.g., "Ruby on Rails" stamped on a delivery driver posting). Tuned conservatively to keep legitimate skills: an independent multi-judge audit confirms about 99% of removed tags are false positives, with fewer than 1 in 500 legitimate skills affected. In production this concentrates the average from roughly 10 to roughly 8 skill tags per posting, focusing each record on the skills that actually define the role.
New Website Launch
Launched the new Canaria website with dataset catalog, methodology documentation, data schema explorer, provider comparison, and solutions pages. Built on Next.js with a modern dark theme.
Automated Sample Generation Pipeline
Launched AI-powered sample generation. When you request a sample, our AI assistant builds an optimized query based on your use case, executes it, reviews quality, and delivers a CSV with signed download link via email. Up to 3 quality review iterations per sample.
Skills Taxonomy Expanded to 40,000+
Expanded the skills taxonomy to 40,000+ technical skills, 3,000+ certifications, and 250+ soft skills. Added a title-aware relevance filter to reduce false positives (e.g., "Java" correctly excluded from Barista postings).
Salary Prediction Model v2
Retrained the salary prediction model on 50M+ Glassdoor/Indeed observations. Coverage for 2023+ data now reaches 85-95%.
Canaria Job Intelligence Platform
Official launch of the Canaria Job Intelligence Platform with 1B+ unique deduplicated job postings, 100+ enriched fields, and our NLP enrichment pipeline. Coverage from 2022 to present with daily incremental updates.
Ready to Explore the Data?
Get a free sample of our research-grade job market data. No credit card required.
Request a Free Sample