Business
Compare ecological investments
Which project or investment fits our ecological goals and budget? Compare declared costs, constraints and the evidence behind each option.
Compare investments →The ecological OS for business, industry & conservation
Decisions that help the planet recover.
Viridis is the ecological operating system for business, industry and conservation. We turn open, machine-checked science into decisions about where to invest, how to operate within ecological limits, and what to protect and restore. Start with a ready service, from corridor priority to wildfire carbon recovery, or bring us the larger decision and we’ll scope it with you.
Free fit check · price agreed before paid work · unknown inputs welcome
What you take away
Example structureA clearer choice.
A record of why.
Should we protect habitat, restore a corridor, or combine both?
- Compare the options
- Costs, ecological effects, and doing nothing.
- See who bears the consequences
- People, places, and effects beyond the project boundary.
- Know what still needs an answer
- Missing evidence, limits, and accountable review.
Prefer to talk it through? Ask the Viridis Guide
Viridis Guide
Tell me the decision you have to make.
Routes to a theorem-backed method · never invents a number
Nothing you type is stored unless you choose to share it. We count which kinds of questions we can't answer yet.
Use public or non-sensitive information only. The Guide routes and explains; the kernels compute; a person reviews before anything is delivered. How it’s built · Run kernels yourself
A real result, not a mock-up
Which corridor link should we protect first?
Two habitat cores are joined by two routes of two links each. Every link has an even chance of holding. One link costs twice as much to protect.
The service computes the exact chance the cores stay connected: 43.75%. It also computes how much each link matters, which a proven theorem shows is that link’s effect on connectivity. Here every link matters equally, so per dollar you protect the cheaper links first.
Exact when links fail independently. Correlated losses such as fire or drought are outside this method, and it says so on every result.
Example output
43.75%
chance the two cores stay connected
| Link | Holds | Cost | Importance | Protect |
|---|---|---|---|---|
| Link 1 | 50% | 1 | 37.5% | #1 |
| Link 2 | 50% | 2 | 37.5% | last |
| Link 3 | 50% | 1 | 37.5% | #2 |
| Link 4 | 50% | 1 | 37.5% | #3 |
From the service’s published example, recomputed on every build.
Questions you can answer today
Proven methods for real conservation decisions.
Each service runs free in your browser and shows its assumptions next to the result. When the answer matters, ask for a reviewed analysis of your own case.
Which corridor link should we protect first?
Exact odds that two habitat cores stay connected, and which link matters most, per dollar if you add costs.
Try it free →After a wildfire, can this forest’s carbon recover on its own?
A proven threshold on four retention and regrowth coefficients tells you which side of recovery a stand is on.
Try it free →Can our nurseries meet planting demand?
The largest share of demand your supply network can fill, and exactly which sites are the bottleneck.
Try it free →How large must a planting cluster be to spread on its own?
The critical cluster size, with a go / no-go for the cluster you plan.
Try it free →Does our parcel ranking hold if the weights are disputed?
A sharp bound on how far scores can move, and whether two parcels can swap places.
Try it free →Has this monitored signal really changed?
An alarm whose false-alarm rate stays below the level you set, however long you watch.
Try it free →Or start from your own question
One question. An inspectable decision map.
This map starts empty on purpose: nothing is assumed until you supply it. Open the workspace to add your conditions, compare alternatives and export the record.
Defines
Checks applicability and records the foundational constraint
Composes through an explicit contract; no automatic derivation
Compares supported conditions
Formal dependenciesEmpirical/model assumptionsUser informationHypothetical valuesUnknownsAlternatives
Start with the work you need to do
Three contexts. One ecological decision workflow.
Choose a starting question, then adapt it to your project. A fit check identifies supported analysis and evidence gaps before you commit to paid work.
Industry
Evaluate operating choices
How should we allocate resources while respecting ecological constraints? Frame alternatives, resource limits and evidence needed for a supported comparison.
Evaluate operations →Conservation
Prioritize protection and restoration
Which actions fit our conservation objectives and available funding? Compare protection, restoration and mixed management under stated assumptions.
Prioritize conservation →These are starting questions, not prevalidated sector models. Method fit and project evidence determine which conclusions Viridis can support.
Choose how to work with Viridis
Work with your team or connect an agent.
For decision teams
Get a reviewed analysis for your decision.
Bring your options, ecological goals, budget and evidence. We agree the scope, fee and delivery date, then prepare a comparison with human review and a report your team can use.
Free fit check · individually quoted analysis
For agents
Build a structured decision report.
Connect your agent, check that a method fits, and approve the exact quote. Receive up to three scenarios through one supported method, with inputs, limits and computation receipts attached.
Free previews · 1 USDC per automated report
Automated reports are unreviewed. Human review is available as a separately quoted service.
The practical product
Understand the choices, the assumptions, and what would change the decision.
A shared decision frame
Bring the question, ecological objectives, economic constraints and alternatives together. Inspect how the evidence connects to the choice.
Ecological and economic tradeoffs
Compare protection, restoration and mixed management against declared budgets. Add ecological impacts, affected communities and proposed limits in a linked account; keep missing evidence unknown.
A portable decision record
Save a scenario on your device or export a readable report and JSON. Inputs, assumptions, source references and kernel versions stay attached.
Today’s workflow supports conditional budget pathways, a linked ecological impact account and an optional resource allocation model. Attach reviewed analyses for weight sensitivity, staged commitment and price conventions, each with its own explicit premises. It does not infer habitat benefits, species responses or an optimal site intervention.
Proven in the open
Check our work. That is the point.
Open proofs
The mathematics behind each service is checked by the Lean 4 proof assistant and published with DOIs, so anyone can inspect it.
Browse the open canon ↗Eco-Lean Bench
90 of our ecology theorems, packaged as a public benchmark for AI theorem provers.
See the benchmark ↗Recompute anything
Every preview and certificate says how to check it and run your own, with open-source code.
Open the code ↗A proof shows the method’s conclusion follows from its stated assumptions. Field inputs and ecological outcomes still need evidence and review.
Explore the research behind the OS
From research to your decision
Research → Ecological OS → decisions
01
Viridis Research
Develops and publishes the scientific framework. Established knowledge, modeling assumptions and formal artifacts feed candidate research; acceptance requires review.
Explore the open science →02
Viridis Ecological OS
Integrates approved, versioned decision kernels. It checks applicability, composes constraints and preserves explanations for inspection.
Inspect the platform →03
Ecological decision service
Viridis Conservation brings that research into an ecological OS for business, industry and conservation: scenario workflows, evidence review, reports and supported operation. A scoped engagement connects the research to your decision and its limits.
Request a reviewed analysis →Looking for land stewardship or community conservation? Visit VergeCommon , our separate destination for that work.
An actual architectural root
Our working premise is the Intelligence Bound conjecture.
Every decision records its root-kernel assessment. Quantitative constraints apply only when the conjecture’s quantities, units and scientific mapping are justified. Other kernels connect through explicit contracts; they do not all follow from this conjecture.
Read the statement, assumptions and current limits →Formal verification and field validation are different.
Lean establishes that a formal conclusion follows from stated hypotheses. Implementation checks assess code and computation. Neither proves field truth, ecological validity, or the correctness of a human objective.
The current root assessment is conditional. A source-unit correspondence review and a justified project mapping are required before a numerical Intelligence Bound result can be claimed.
Open science. Usable decision systems.
The mathematics is public. The product makes it useful.
Papers, formal artifacts and reference software are published through GitHub and Zenodo according to each artifact’s license. The nightly engine proposes scientific connections and candidate kernels from established knowledge. It is not a field sensor network, and candidate output cannot silently change production.
Customers pay for integrated workflows, scenario management, transparent reports, hosted operation, deployment, maintenance, support and domain-specific packaging. The commercial application and customer systems retain their existing product and licensing boundaries.
Project quotes show the scope, fee, payment milestones and acceptance criteria before work. Accepted quotes keep their agreed price; changes require a new review. Existing specialist offers remain available.
The long-term mission
Software for the healing epoch.
Viridis builds ecological software for the healing epoch: decisions that help land, water and life recover, proven in the open. We start with the choices conservation teams make every season, and we publish the mathematics so anyone can check it, build on it and hold us to it.
AI alignment is a related research motivation. Physical constraints do not select human values or conservation objectives. Decisions remain with accountable decision-makers.
Viridis does not issue environmental credits, guarantee conservation outcomes, or authorize autonomous ecological intervention. A Decision Chain preserves the evidence and limits for human review.
Ready when you are
Bring the decision you actually have to make.
Two minutes to a suggested method and the evidence it needs. Free, no account. Your draft stays in this tab; a fit check does not submit a service request.