Every energy and climate researcher in Maryland, in one place.

Today that expertise sits across 28 university websites, each in its own format, with no way to search between them. This is one searchable roster, built for MEIA staff and the founders they work with. Below is a working demo on real University of Maryland data, the method behind it, and the proposal to build it out.

01 · The prototype

Describe a venture. Meet its researchers.

Live demo on 14 energy-research faculty from UMD's Maryland Energy Innovation Institute, the sample slice of a ~90-person roster. Type anything, or start from an example.

Try:
Entrepreneur queue · demo
Three faces of one engine, running entirely in your browser on 14 of the 91 UMD faculty: founders search it, MEIA staff get a vetted match memo from it, and anyone can browse the inventory underneath. The production system matches on meaning: a founder who writes "storing wind power overnight" finds the flow-battery lab that never uses those words. It covers every school in scope, and keeps a human in the loop: every match carries its reasoning so MEIA staff can vet it before anyone is introduced.
02 · The landscape

The whole state, surveyed

Every four-year institution in Maryland, checked for energy and green-tech research activity.

28
four-year institutions statewide
12
with active energy / green-tech research
186
researchers already verified by name
InstitutionEnergy researchResearchers
UMD College Park R1 Maryland Energy Innovation Institute (MEI²), CEEE thermal-systems center, CALCE battery reliability 91 verified
Johns Hopkins R1 Ralph O'Connor Sustainable Energy Institute: grid decarbonization, offshore wind, net-zero policy; includes APL 95 verified
UMBC R1 AMEE lab (batteries, fuel cells, hydrogen), thermal-fluids group, grid modeling, NASA GESTAR II climate consortium ~20–30 est.
UMCES Statewide environmental-science labs; $11M offshore-wind research partnership with US Wind ~15–25 est.
Morgan State R2 · HBCU CAESECT energy-systems center (fuel cells, solar, wind, biomass), DOE urban-climate lab, NSF CREST materials ~10–15 est.
8 more institutions Frostburg's smart-grid cluster, Naval Academy's energy-security center (2024), Hood's endowed biofuels chair, Towson, UMES + three more HBCUs ~15–30 est.
Statewide Energy-tech core (roughly doubles counting climate science & energy policy) ~230–290

Verified = counted by name from institute directories, July 2026. Estimates are roster-based; the Phase-1 extraction pass replaces them with per-person classifications. Community colleges (16) excluded: workforce training, not research.

03 · Method

How the data is built, and why you can trust it

Collect

We collect the public faculty pages from each university's website, carefully and within each site's terms.

Normalize

Every school lays out its pages differently. We read them all into one consistent format, and never add anything a page doesn't actually say.

Shared taxonomy

Free-text research blurbs map to a controlled vocabulary of research areas, tuned with MEIA's domain knowledge, so "solid-state batteries" means the same thing at UMD, Hopkins, and UMBC. This is what makes research comparable and matchable across schools.

Match

You describe what you're building in plain words. Instead of matching keywords, the system finds the researchers whose work actually fits what you mean, ranks them, and tells you in plain English why each one came up.

Deliver

A web tool MEIA brands and links from its site, plus a one-page match memo per entrepreneur for staff-brokered introductions. Runs on inexpensive managed infrastructure; the data and results are MEIA's.

Provenance on every row

Each researcher entry links back to the source page it was read from, with a fetch date, so any claim can be checked in one click.

Accuracy, sampled and scored

Before we hand anything over, we spot-check a random sample against the original university pages and report exactly how accurate it is, so you never have to just take our word for it.

Contact info, handled deliberately

All data comes from public university pages, but the public tool shows profile links only; direct contact details stay in MEIA's staff view until the opt-in question is settled.

Freshness you can see

Faculty rosters drift. Every page shows when it was last crawled, and the refresh option below keeps the inventory living instead of aging into a snapshot.

04 · Proposal

Research Match, statewide

Every four-year institution in Maryland with energy or climate research, all four HBCUs included. About 250 to 450 researchers, with two full years of support.

$27,500all-in
Build: the inventory, the shared taxonomy, the matching engine, the public match page, and the staff memo workflow$21,500
Support and updates, two full years$6,000
Total$27,500

After Year 2, $3,000 a year to keep it live, optional. If the budget needs to be smaller, coverage can narrow to the anchor institutions (UMD, Johns Hopkins, and UMCES) for $18,500.

The build runs in three stages:

01

Match engine, one school

The reference build everything else scales from: UMD, end to end.

  • Full ~90-faculty MEI² roster, structured on the locked schema
  • Research-area taxonomy defined with your team
  • Production matching with plain-English reasoning on every result
  • The founder-facing match page and the staff match memo with draft introduction
02

Multi-school rollout

Additional universities on the same pipeline: Johns Hopkins (ROSEI, 95 researchers verified), UMCES, UMBC, and Morgan State, normalized to the shared taxonomy so the match runs statewide, not school-by-school. The remaining small institutions (Frostburg, Towson, USNA, UMES, Hood, and peers, ~15–30 researchers combined) come as one bundle rather than school-by-school.

03

Support & updates 2 years included

Two full years of monthly re-crawls, hosting, and match-quality review, so the inventory tracks faculty as they move and the matches stay sharp, instead of aging into a one-time snapshot. Included in every option above.

One fixed priceNo retainer lock-in, no open-ended meter. You know the full cost before we start.
Billing50% to begin, 50% on delivery.
You own the outputThe data, the taxonomy, and the delivered results are MEIA's.
Human in the loopThe system proposes; MEIA disposes. Nothing reaches a researcher or founder without your sign-off.
For the state grant application

If this is going into the proposal, we'll draft the technology section for you: a technical narrative, a line-item budget, and this working demo to cite as preliminary work.

Your tech section, drafted

Narrative, line-item budget, and the landscape survey, ready to paste into the application.

You file stronger

The application cites a working prototype and a verified statewide survey as preliminary work.

Phase 1 starts

Full UMD roster, taxonomy, production matching. Live in ~3 weeks.

Reporting built in

Every match, introduction, and outcome logged. Your grant compliance numbers, automatic.

05 · Team

Who's building it

Natan Lawson
Natan Lawson
Builds & delivers

Natan Lawson runs Bike Powered Events and builds the software behind it. Across seventeen years of running his own companies he has built the systems they depend on, learning to build because the software he needed did not exist.

Julian Coy
Julian Coy
Technical advisor

Julian Coy is a software engineer and community organizer in Baltimore. He builds large software systems and advises Worktree on technical architecture, code quality, and long-term maintainability. He also organizes the Code Collective community and its regional calendar, connecting Baltimore's civic tech, startup, and developer scenes.

06 · Open questions

Five decisions that shape the build

These are the calls we'd make together in the first Phase-1 working session. Answers to any of them now sharpen the scope.