A decision layer for supply chains under pressure

Discover your
exposure before
the next disruption
hits.

Your supply chain data already exists, scattered across six systems that don't talk to each other. Map Collective fuses what you already have — bills of materials, trade records, vessel movements — into one live map of where you're exposed and how far a shock travels. No survey. No waiting on suppliers to self-report. The picture you needed before the headline broke.

6 silos
hold fragments of the one question none can answer alone
~60%
of global trade sits below the declared customs layer
n-tier
cascade depth traced from a single supplier or chokepoint shock
60 days
to your first live exposure map — one product line or input
01 · The problem

Six teams. Six silos. None of them can answer the one question that matters.

“What is our actual exposure, right now — and what should we do about it before the next disruption hits?”

Four of these silos are a door into the same exposure engine — cascade risk, compliance, tariffs, origin & forced labor. Open whichever one is on your desk today.

~60%
of global trade activity sits below the declared customs layer — unmapped, unmodeled, untraced.
30–50%
typical response rate on quarterly supplier surveys — the industry's current best instrument.
02 · How GRID works

GRID turns the data you already own into a live read on exposure.

01 · Integrate

Unify your data where it lives.

One connected graph across four source layers — internal ERP and BOM; external bill-of-lading, AIS, sanctions, ownership; federated queries against willing suppliers in-jurisdiction; and dark-node inference where data is missing. No central pool. No breach surface. No migration.

ERP / BOM Bill of lading AIS & ownership Federated query Dark-node inference
02 · Prioritize

Surface what actually matters.

A unified graph is still noise without ranking. GRID orders signal by business impact — production days lost, revenue at risk, material shortfalls — not arrival order. Your highest-exposure nodes surface first; everything else stays out of the way until it isn't noise anymore.

Impact-ranked views Real-time Daily operating interface
03 · Decide

Answer the actual question.

30/60/90-day cascade forecasts. Scenario modeling for supplier loss, chokepoint closure, sanctions shock. Alternative supply chain designs scored across cost, risk, and time. Every output confidence-scored. Every data point provenance-tracked. Audit-ready by construction.

Cascade forecast Scenario modeling Decision scoring Provenance graph
03 · What the decision layer looks like

Three views into one living graph. A single source of truth underneath.

We don't ask your suppliers to fill out a form, and we don't guess at what we can't see. We fuse the data that already exists into one graph — then read your exposure off it.

01 · Dependency graph Live

N-tier supplier graph, surfaced from BOM.

Materials traced root-to-canopy across tiers, with confidence and exposure on every edge. No survey required.

7 tierstraced
2,418edges
94%confidence
02 · Logistics intelligence Live AIS

Vessel flows along the lanes that move your inputs.

Bill-of-lading + AIS fused. Lanes, ports, and chokepoints monitored continuously — arrivals surfaced before they're surprises.

37lanes
142vessels
6chokepoints
03 · Risk intelligence Updated 14:32 UTC

Geopolitical, climate, and conflict overlays on the dependency canopy.

Forced-labor zones, geopolitical risk, shipping chokepoints, and active disasters — layered, scored, and routed back to the exposure that matters.

4medium risk
0high risk
22overlays
04 · Declared vs. observed

What's declared, and what's actually happening. We show you both — and where they part.

A name on a form is just a string. We lay what's declared against what the world's transactional signals show, and label every node for exactly what it is.

Corroborated Declared, and confirmed in independent transactional data. Two views agree. Both graphs
Additional A node the declared graph never mentioned, surfaced from observed activity. Transactional only
Divergent Observed activity contradicts what was declared. The gap is itself the signal worth investigating. Conflict
Unobserved Declared, but not yet seen in external data. Reported plainly as unobserved — never inflated into proof. Declared only
Enriched Every node carries its own dossier: ownership, geography, trade activity, sanctions, risk. Per node

We sell you synthesis you can act on — never a guess dressed up as certainty.

05 · Architecture

Your data stays where it lives. GRID integrates and decides at the edge.

01

No data migration. Integrates in place.

Your ERP stays in your ERP. No central lake to build, no warehouse to maintain. GRID reads where data already is and writes back as decisions.

02

Enterprise and government ready, by design.

Deployable inside your data perimeter. SOC 2 Type 2 certified. Compatible with data-sovereign environments and classified networks — the same platform serves both.

03

Gap-fills without surveys. Compounds with use.

Where supplier data is missing or withheld, GRID infers — confidence-scored at every step. Every deployment enriches the shared dependency graph the next one inherits.

04

Six categories of proprietary agents.

Scraper backends. LLM provider routing. Data enrichment. Disaster polling. Risk signal generation. And the recommendation layer — the one that closes the loop from signal to action.

06 · Why now

We saw the chokepoint close before it was a crisis.

When the Strait of Hormuz seized, the companies that moved first weren't the ones with the most dashboards — they were the ones who already knew which of their inputs ran through it.

Map Collective mapped that exposure ahead of the disruption. The next chokepoint is already forming. The only question is whether you see your exposure before it's on the front page — or after.

Live triggers, each a door into the exposure engine: a compliance deadline · a tariff shock · a supplier failure.

Want to see the engine on public data first? Explore Planet → Particle. This is public data — we do this on your supply chain, privately, to the part number.

07 · What GRID replaces

Not a new data vendor. A replacement for the processes that are failing your teams.

Decision need
Today's process
With GRID
Sub-tier supplier mapping
Consultant-led mapping exercise: 3–6 months, $500K–$2M, out of date on delivery.
41,300+ supplier relationships mapped in days, validated against ground-truth BOM. No survey required.
Supplier risk intelligence
Quarterly supplier surveys at 30–50% response rates. Stale within weeks. No confidence scoring.
Continuous, confidence-scored risk signals across all tiers, ranked by business impact, updated in real time.
Disruption scenario planning
Static risk reports or reactive war rooms fired after disruption has already begun.
On-demand 30/60/90-day cascade modeling. Supplier loss, chokepoint closure, sanctions shock — production days lost, revenue at risk.
Compliance traceability
Manual filings, supplier self-attestation, legal review — months of effort per submission.
N-tier origin tracing with Proof-of-Authority provenance. Audit-ready for UFLPA, CSRD, EUDR, Section 1260H — without supplier opt-in.
Data integration & unification
Six-month internal data lake initiative, systems integrator engagement, ongoing maintenance burden.
One unified graph across ERP, BOM, trade, logistics, risk. No migration, no central lake, no new infrastructure project.
08 · Built by operators

Engineers, supply chain operators, defense practitioners, and domain researchers — one team.

Tara Gupta
Founder & CEO
Forbes 30 Under 30 (Energy). Georgetown MBA, RISD BFA. Principal Investigator on four NSF SBIR awards. Six years from concept to revenue to enterprise deployment.
Isaac Hicks
CTO & AI Architect
Architected GRID's data integration and decision platform. Led live enterprise deployments mapping tens of thousands of supplier relationships at OEM scale. Designed the confidence and provenance architecture.
Marco Ruiz
Principal AI Engineer
ML systems architecture. Leads the probabilistic data integration pipeline, cross-source signal fusion, and the confidence scoring engine underpinning every output.
Advisors
Prof. Volodymyr Babich
Research
Georgetown McDonough. 66+ peer-reviewed publications. NSF-funded multi-tier dependency modeling. AI/ML supply chain research.
Kate Jones
Geopolitical & quantum
MS Quantum Computing & Cybersecurity, Oxford. Pentagon / White House geopolitical strategist. Sparse data interpolation architecture.
09 · Start

Find your exposure in 60 days.

A scoped pilot on one product line or one critical input. We surface your N-tier dependency, fuse it with live logistics and risk, and hand you the exposure picture your six teams can't assemble — in weeks, not quarters.

What you'll have at day 60
  • Your N-tier dependency map, surfaced from BOM — no survey.
  • Live logistics and risk overlaid on the inputs that matter.
  • Exposure ranked by how far a shock actually travels.
  • Every node labeled — corroborated, divergent, or unobserved.
  • The exposure picture your six teams can't assemble alone.