Open Scholarships

A free, open directory of scholarships and student aid — Nevada first — published as machine-readable data anyone can use, with a thin public read API on top.

Most scholarship listings are locked inside proprietary aggregators (Scholarships.com, Fastweb, Bold.org) or licensed datasets (CareerOneStop's listings are licensed from Gale/Cengage). There is no free, open, machine-readable source. This is that source. The data is the product; the API is just a tap. The goal isn't reach — it's to be the authoritative, current, openly-licensed record of the aid a Nevada student can actually get, and to grow outward from there.

License: data is CC BY 4.0 (use it freely, just attribute). Code is MIT.

Live: scholarships.grudged.io — docs + full dataset at /scholarships.json, query API at /api/scholarships?state=NV&level=high-school-senior. CDN-served, CORS-open, no key. Auto-published from this repo on every push (Netlify static + edge function); the FastAPI service in api/ is the internal curation/dev engine.

Why it's built this way

• Curation is the value. A stale scholarship DB is worse than none — a kid who chases a dead link or a passed deadline is harmed. Every record carries provenance (source_url, last_verified, method) so trust is verifiable, and a freshness checker flags rot.
• Facts from primary sources only. Records are gathered from sponsors' own pages, .gov, and .edu — never copied from a proprietary aggregator's compiled list.
• Local-authoritative beats national-shallow. Own Nevada completely first (state programs + every NSHE institution + Nevada/Vegas foundations + civic awards), then expand by state.

Data model

One JSON file per scholarship under data/<STATE>/, validated against schema/scholarship.schema.json. Key fields: name, sponsor, type, award (amount/basis/renewable), deadline, eligibility (residency, level, fields, GPA, tags), geo, links, provenance, and status. New records start needs-review; only active records are served by default.

API

GET /healthz                      liveness + record count
GET /meta                         license, attribution, counts, states covered
GET /scholarships                 filters: state, level, field, basis, type, sponsor_type,
                                   deadline_after, amount_min, q, status, limit, offset
GET /scholarships/{id}            one record

Every response carries X-License / X-Attribution headers; CORS is open (any site may read it).

Run locally

cd api
python3 -m venv venv && ./venv/bin/pip install -r requirements.txt
./venv/bin/uvicorn app.main:app --reload --port 8930
# http://localhost:8930/meta  •  /docs for the OpenAPI explorer

Maintain the data

python tools/extract.py <official-url> --source-name "..."   # Gemma-assisted draft -> drafts/
python tools/check_freshness.py [--write]                    # flag dead links / passed deadlines
api/venv/bin/python -m pytest tests -q                       # validate every record vs schema

Status

v0.1 — Nevada seed (Governor Guinn Millennium, Silver State Opportunity Grant, Nevada Promise), all needs-review pending source verification. First consumer: FirstEmbark.

Open Scholarships by Grudged LLC.