AquaSight

AquaSight is a groundwater exploration pre-feasibility intelligence product executed through the envAIros platform. It’s designed for early-stage decisions—when you need a defensible first pass on whether groundwater exploration is justified at a site, what the likely constraints are, and what should happen next.

AquaSight is built on a hybrid foundation: scientific context + deterministic decision logic, with AI used only to explain and structure reporting after deterministic outputs are produced.

AquaSight is built for teams that need a consistent screening output before committing to field spend:

• Environmental consultants and hydrogeologists doing early feasibility screening

• Developers, landowners, and asset teams needing a rapid first-pass assessment

• Project managers who need consistent, repeatable screening outputs for reporting and stakeholder alignment

AquaSight combines site context and deterministic screening logic to produce a structured outcome and clear rationale:

• Builds site context relevant to groundwater feasibility (location-based context pack)

• Applies deterministic screening logic to produce a consistent classification and rationale (repeatable outcomes for the same inputs)

• Generates report-ready outputs including constraints, drivers, assumptions, and recommended next steps

• Uses AI to explain—not decide: AI may explain decisions and summarise risks/assumptions, but may never influence scoring or override feasibility logic

All Sight products execute through a fixed pipeline where AI is positioned after decisions are derived:

1. Input capture & validation

2. Data retrieval & versioning

3. Scientific context construction

4. Deterministic models / rules evaluation

5. Scoring & decision classification

6. AI explanation & structured reporting

• Site location (latitude/longitude)

• Intended use (e.g., domestic, agricultural, industrial) and demand band

• Basic site notes (optional): constraints, land use, access, existing boreholes

AquaSight is designed to be auditable and traceable. Every output must be traceable to datasets, rules, models, assumptions, and limitations.

Core outputs:

• Deterministic risk-based classification (canonical system meaning)

• Key constraints and drivers (why the classification was reached)

• Recommended next steps (what to verify, what to collect, what to test)

• Report-ready summary suitable for stakeholder communication

• Trace metadata (e.g., ruleset/dataset version references where applicable) to support auditability

• “Can this property likely support a borehole for domestic supply?”

• “Is this site worth investing in drilling and pump testing?”

• “What are the major feasibility constraints I should flag to the client?”

envAIros is designed so decisions remain defensible under scrutiny:

• AI explains. AI never decides

• AI may explain and summarise, but must never generate scores, alter classifications, or override deterministic logic

• Outputs are designed to be traceable to the inputs, rules, and assumptions that produced them

• Identical inputs + identical ruleset version must yield identical outputs (determinism guarantee)

• Not professional sign-off

• Not a regulatory approval engine

• Not an autonomous decision-maker
  envAIros delivers decision intelligence, not guarantees or certification.