Awards by year
Filter-aware preserved-row counts.
Funding horizons
Major grant calls
Starting groups, remote labs, consolidation and major research cohorts.
● Starting groups● Remote labs● Consolidation● ResearchDashed bars = cohort estimates · drag the time axis to pan
Newest cohorts first. Funding totals use the explorer’s display normalization of annex units; they are indicative, not authoritative budgets.
Dates, funding and sources
Bars show estimated cohort schedules, inferred from a cited funded project or institution report, rather than verified dates for every award. Months include the published end month. Decision dates can differ from project starts. No extensions or early terminations are assumed. Counts and funding use the full audited dataset and stay fixed as table filters change. Funding is the sum of numeric amounts in the result annexes, not annual spending; partial coverage is stated. An unavailable total is never shown as zero.
Find an award
Combine filters; the view and URL update immediately.
Tip: click a chart bar to filter.
| Status | Source |
|---|
No records match these filters.
Host institutions
Complete filter-aware list of normalized institutions.
Smaller institutions are grouped as Other NAS or Other non-NAS. Set the threshold to 1 to show every institution separately.
How to read this explorer
- View raw record preserves either the original extracted text or the rendered-page audit note. Normalized fields for re-audited rows reflect the visual table.
- Shortlisted reproduces the research audit's title-based AI/ML shortlist; keyword candidate is a broader automated flag, not a substantive classification.
- “Not stated” means not reliably recoverable from the public annex, not that the underlying value is zero.
Funding and extraction caveats
Funding values are shown exactly as recorded in the audited appendix. Broad range filters use a display-only normalization heuristic because annexes mix thousand-AMD and whole-AMD notation. They should not be used to calculate an authoritative total.
Generated IDs preserve genuinely uncoded rows. Ten annex tables with unreliable reading order were re-audited from rendered official pages with Gemini 3.1 Pro. The 2023 equipment table's vertically merged organization cells and the 2026 professional-training table's same-line PI/title boundaries were mapped by visual checks.