Datasets I created — including synthetic ones like ClaimWise — plus a few useful third-party sets I keep copies of, clearly credited on each row. Most open directly in theData Studio. Catalog:datasets.parquet.
The AI agents behind /agents — purpose, framework, links and creation date for each.
parquet3 KBmineJob Scout hiring-trends snapshot across ~95 companies' ATS boards (refreshed daily).
parquet571 KBmineClaimWise synthetic healthcare RCM — activities (largest seed).
csv144 KBmineAll seven ClaimWise seed CSVs (users, facilities, health plans, collections + above).
csv213 KBminePublic container images on ghcr and Docker Hub, with tags and pull counts.
parquet3 KBmineFollower and content-viewer demographics by seniority, industry, job title, company, size and location — one row per snapshot, split by demographic_kind.
parquet5 KBmineDaily LinkedIn impressions, engagements and new followers; upserted across snapshots.
parquet4 KBmineOne row per LinkedIn analytics export: period, impressions, members reached, follower total.
parquet1 KBmineSource CSV behind linkedin-analytics-top-posts.parquet.
csv17 KBmineTop LinkedIn posts with impressions and engagements, joined to this site where the post is known.
parquet12 KBmineSource CSV behind posts.parquet — LinkedIn articles, full export columns.
csv65 KBmineSource CSV behind linkedin-posts.parquet — standalone posts with engagement.
csv54 KBmineStandalone LinkedIn posts (no article behind them) with full text and impression/reaction/comment counts.
parquet33 KBmineClaude Code skills published on this site, with file counts and download paths.
parquet2 KBmine