Start with the question
you came with.
There is no single correct route through a living system.
For scientists
What must we prove?
Explore functional soil biology, model generalization, causal evidence and the boundary between food composition and health.
Follow this path →PATHWAY 02For builders
What could we build?
Explore the observation-to-decision loop, evidence provenance and a modular intelligence architecture.
Follow this path →PATHWAY 03For investors
Where does value accumulate?
Separate total ecosystem benefits from addressable markets, business models and realistically capturable revenue.
Follow this path →PATHWAY 04For founders
When should a venture exist?
Explore Subzero, staged experiments, kill criteria and the threshold for earned ignition.
Follow this path →For scientists: make the uncertainty visible.
Turn big claims into small, independently testable hypotheses across physical, chemical and biological systems.
- 01
Specify the scientific question and outcome variable.
- 02
Identify known mechanisms and competing explanations.
- 03
Choose measurements and controlled comparisons.
- 04
Publish uncertainty and negative findings.
What would count as evidence?
- Pre-registered hypotheses
- Independent replication
- Open methods and calibration records
What could break the thesis?
- Proxy fails under new conditions
- Causal effect cannot be isolated
Which scientific link is weakest?
Do not assume the answer. Identify the relevant stakeholders, competing explanations, baseline measurement, and a test that could change the decision.
What would convince a skeptical agronomist?
Do not assume the answer. Identify the relevant stakeholders, competing explanations, baseline measurement, and a test that could change the decision.
For builders: connect evidence to action.
An intelligence product must close the loop from observation through decision and measured outcome.
- 01
Capture interoperable observations.
- 02
Model context, uncertainty and provenance.
- 03
Surface a recommendation in the existing workflow.
- 04
Measure actual outcomes and learn.
What would count as evidence?
- Usability tests with real operators
- Baseline-vs-model field performance
- Audit trails for recommendations
What could break the thesis?
- Integration friction blocks use
- Recommendation is not trusted
What is the smallest working end-to-end loop?
Do not assume the answer. Identify the relevant stakeholders, competing explanations, baseline measurement, and a test that could change the decision.
Which integrations matter before AI sophistication?
Do not assume the answer. Identify the relevant stakeholders, competing explanations, baseline measurement, and a test that could change the decision.
For investors: separate possibility from capture.
A trillion-dollar ecosystem claim is not a revenue forecast. Follow who benefits, who pays and who owns the decision.
- 01
Quantify a specific addressable decision.
- 02
Estimate net incremental value.
- 03
Validate price, deployment economics and retention.
- 04
Map competitive and institutional control points.
What would count as evidence?
- Paid pilots
- Unit economics and sensitivity bands
- Competitive alternatives
What could break the thesis?
- High theoretical value but low budget ownership
- No defensible distribution advantage
What makes this venture-scale rather than a service?
Do not assume the answer. Identify the relevant stakeholders, competing explanations, baseline measurement, and a test that could change the decision.
Where could value migrate away from the startup?
Do not assume the answer. Identify the relevant stakeholders, competing explanations, baseline measurement, and a test that could change the decision.
For founders: earn ignition.
Subzero is a protected state for exploration before the commitments of venture formation.
- 01
Name the unknowns.
- 02
Design the smallest falsification test.
- 03
Build a cross-disciplinary founding network.
- 04
Form the venture only after scientific and commercial thresholds are met.
What would count as evidence?
- Clear gate criteria
- Repeatable outcomes
- Committed customers and partners
What could break the thesis?
- Scientific novelty without customer demand
- Business interest without credible science
What evidence is enough to form a company?
Do not assume the answer. Identify the relevant stakeholders, competing explanations, baseline measurement, and a test that could change the decision.
Who is missing from the founding conversation?
Do not assume the answer. Identify the relevant stakeholders, competing explanations, baseline measurement, and a test that could change the decision.