Decision support system2026Prototype

PropOps

A property-diligence system that gathers fragmented records, surfaces inconsistencies, and keeps judgment with the human operator.

Back to projectsRepository

At a glance

Problem

Important property decisions are slowed down by fragmented records and hard-to-interpret legal signals.

Approach

An AI-assisted diligence workflow that gathers, cross-checks, and summarizes risk-relevant signals.

Why it matters

The value is not prediction theatre. It is making evidence easier to inspect.

System sketch

01Gather

Pull registry, legal, and compliance context from multiple fragmented sources into one workflow.

02Normalize

Convert scattered formats and inconsistent records into a shape that supports reasoning instead of manual hunting.

03Surface risk

Highlight missing information, inconsistencies, and open diligence questions that deserve attention.

04Support judgment

Keep the human decision-maker in control while making interpretation faster and more legible.

Design notes

Interpretation is the bottleneck

The hard part is often not access to data. It is turning scattered evidence into something usable.

AI should reduce ambiguity

The system is more useful when it clarifies uncertainty than when it performs confidence theatre.

Practical problems matter

Some of the most interesting systems work lives in ordinary decisions with real consequences.

Question

Property diligence often breaks down at the interpretation layer. The information exists, but it is fragmented, inconsistent, and slow to reason through under time pressure.

PropOps reduces that burden without pretending the decision itself can be automated away.

Approach

  • Cross-reference registry, compliance, and legal signals across multiple sources.
  • Surface missing information, inconsistencies, and open diligence questions explicitly.
  • Use AI as a due-diligence assistant rather than as an oracle or a risk score generator.

Open questions

  • How should uncertainty be represented without overwhelming the user?
  • What evidence model makes manual review faster instead of simply generating another summary layer?
  • Where should the boundary sit between automated synthesis and explicit human judgment?

Project spec

RoleProblem framing, systems design, and implementation.
StatusPrototype
TypeDecision support system
UpdatedApr 2026
Primary objectA diligence trace built from evidence, inconsistencies, and open questions.
System boundaryRetrieval, normalization, and risk surfacing across fragmented records.
Current artifactNarrow property-diligence prototype.
Pressure pointRepresenting uncertainty without hiding the underlying evidence.

Topics

decision supportretrievalrisk analysisinformation synthesis

Where it is now

  • The current prototype is intentionally narrow: diligence before commitment, not end-to-end transaction workflow.
  • Most of the difficulty is in normalization, source traceability, and uncertainty handling rather than summary generation.