Don't just change jobs. Change your coordinates.AetharaEmployer-direct hiring intelligence

Transparency

Changelog & data methodology

Aethara is an independent research product. We pull live postings directly from employer ATS boards (Greenhouse, Lever, Ashby), de-duplicate them to one listing per requisition, and score each with a published Hiring Aggressiveness scoring. Apply links route to each employer’s official careers page.

1,635De-duplicated live roles
246Employers
Jun 24, 2026, 12:46 AM UTCLast refreshed

How we collect & score

Hiring Aggressiveness (HA) estimates how hard an employer is hiring for each specific opening (0–100). Scores combine company-level hiring push with role-specific signals like posting freshness, ATS recency, and how many sibling seats are open. Every score includes a confidence read so thin or incomplete feeds are flagged rather than overstated. Bands: 85+ Hypergrowth, 65+ High, 45+ Moderate, below 45 Minimal. Roles older than 30 days (posted or first-seen) are removed automatically.

Release log

  • DataOne ATS requisition = one listing

    Reqs that an employer posts to several states or offices under a single ATS job id are now collapsed to one canonical listing with a coverage list, instead of one row per location. This removes phantom per-state duplicates (e.g. a remote role cloned across 49 states) and corrects the catalog count, the employer rankings, and per-state page contents.

  • DataDeterministic geo bucketing

    Every role’s US state is resolved from its apply-id state suffix and reconciled against its bucket, so a role can no longer appear under a state that contradicts its location. Non-US roles are dropped at ingest.

  • MethodologyOne HA version, one band vocabulary

    Hiring Aggressiveness is published as a single score per role. Company hiring push and role-specific urgency are combined with a confidence read. The map vocabulary (Aggressive/Active/Steady/Selective) maps one-to-one onto Hypergrowth, High, Moderate, and Minimal bands shared by badges, filters, and the 3D map.

  • MethodologyConfidence shrinks thin-data scores

    A 0–100 confidence value reflects feed completeness and sample size, and pulls thin-data scores toward 50 so a single sparse listing cannot read as a high-intensity market.