alignprime/nevada-innovation-index

Nevada Innovation Index

A structured, source-linked index of 25 Southern Nevada organisations and 33 open funding streams and programmes — venture funds, state offices, university innovation hubs, accelerators, small-business advisers, and local builder communities — each tagged against a fixed 51-subject taxonomy.

Published by Align Prime, a public record of what people are building.

What this is for

Finding the capital and infrastructure that exists in a specific region is a research problem, not a search problem. A founder in Las Vegas looking for pre-seed capital does not need ten million results; they need the eleven things that actually apply to them, each one saying why.

This dataset is the raw material for that. Every row carries the subjects it belongs to, so a project described in plain language can be matched against it by simple set operations — no embeddings, no model, no inference cost.

Nothing here is verified

Nothing in this dataset is verified by Align Prime unless a record carries verified = true. Every row is sourced from the organisation's own public listing. Being listed here is not endorsement, membership, or any agreement by that organisation.

In this release, verified is false on all 25 organisations — including the U.S. Small Business Administration and the National Science Foundation. That is not an oversight. Nobody has checked them in the sense that word is going to mean here, so the field says so.

listing = "listed" means exactly that: the organisation is listed as a public resource. It has not joined anything and has not agreed to anything.

Descriptions are written from each organisation's own published material and say so in their own text ("Organised by Tech Alley, not by Align Prime — check their site before turning up."). Where a figure was not published, the record says it was not published rather than estimating it.

Fame-blind by construction

There is no popularity signal in this dataset. No follower counts, no engagement scores, no trending rank, no editorial ordering.

Relevance is computed from subject rarity: how selective the shared subject is across the whole graph. A subject carried by three records is strong evidence of a match; one carried by two hundred is weak. A large national body and a nine-person run club are ranked by the same rule, and the narrow match outranks the famous one.

That is why matches on the live site read "Matched on finance · Rarity 2/3" rather than a similarity score — the reasoning is inspectable, and you can disagree with it.

Schema

organizations (25 rows)

field	type	notes
handle	string	stable key; resolves at align-prime.com/o/{handle}
name	string	as the organisation publishes it
kind	enum	community · company · government · investor · nonprofit · school
description	string	sourced from the organisation's own listing
website	string	official URL
location	string	nullable
topics	list[string]	taxonomy slugs
verified	bool	false throughout this release
listing	string	listed — a public resource, not a member
listed_at	timestamp	when the record entered the index
align_prime_url	string	canonical page

opportunities (33 rows)

field	type	notes
id	uuid	resolves at align-prime.com/opportunity/{id}
title	string
kind	enum	grant · program · fellowship · scholarship · internship · apprenticeship · mentorship · volunteer
provider	string	the body actually delivering it
description	string
location / remote	string	onsite · remote · hybrid
apply_url	string	where to apply
source_url	string	present on every row — where the claim came from
closes_at	timestamp	nullable; null means no published deadline
posted_at / last_verified_at	timestamp
topics	list[string]	taxonomy slugs

All 33 are open at time of export. Closed and expired records are excluded.

taxonomy (51 subjects)

Fixed, human-curated slugs grouped into clusters (craft, capital, community, …). Small and closed on purpose: a taxonomy that grows with every new record stops being able to say two things are about the same subject.

Files

• dataset.json — everything, including the taxonomy and its clusters.
• nevada_matrix.csv — flat table, one row per record, topics joined with ;.

How it was built

The export authenticates with the same publishable anonymous key that ships inside the public website, and reads only through the security-definer functions that key may call. Both underlying tables refuse it outright:

GET /rest/v1/organizations  ->  401  permission denied
GET /rest/v1/opportunities  ->  401  permission denied

So the privacy property is structural rather than a promise: if an anonymous visitor to align-prime.com cannot already see a field, the exporter cannot read it, and it cannot reach this dataset. Member identities, private projects, unpublished drafts and owner records are not filtered out — they are never fetched. A second independent pass then fails the export if an identity or contact field appears anyway.

The exporter is standard-library Python with no dependencies.

Coverage and limits

• Regional, not national. Southern Nevada. National bodies appear only where they operate a Nevada programme.
• Small. 58 records. It is an index of a specific ecosystem, not a comprehensive funding database.
• A snapshot. Deadlines move and programmes close. Check source_url before acting on any row.
• Unverified. See above. Treat every row as a pointer to a primary source, not as a substitute for it.

Citation

@misc{alignprime_nevada_innovation_index,
  title  = {Nevada Innovation Index},
  author = {Align Prime},
  year   = {2026},
  url    = {https://align-prime.com}
}

Licensed CC BY 4.0. Corrections are welcome and stay on the record.