← Selected work
C-02 Real estate & property

AI-first property-management platform

An AI-first platform that turns a council's years of documents into something you can ask questions of — and drafts the routine correspondence.

Real project · client withheld · outcomes discussed under NDA

  1. 01 Document intake
  2. 02 Grounded chat over the records
  3. 03 A messaging assistant
1730 / C-02 / Document workflow Build path / 3 stages Document ingestion & parsing

Residential strata / HOA councils run on documents — bylaws, minutes, financials, correspondence, maintenance records — that accumulate for years across inboxes, PDFs, and shared drives. The people who need answers are volunteers, not archivists.

The problem

Answering a simple owner question — "what did we decide about this, and when?" — meant digging through years of unstructured records. Routine messaging ate time that council members do not have.

What we did

  1. 01
    Document intake

    A pipeline that ingests and organizes the council's existing records, instead of asking anyone to re-file years of material.

  2. 02
    Grounded chat over the records

    Question-answering that cites the underlying documents, so an answer is checkable against a real record rather than taken on faith.

  3. 03
    A messaging assistant

    Drafts routine correspondence for a human to review and send — assistance, not autopilot.

Relevance

Where this pattern applies.

  • Your team answers repeat questions from years of PDFs, email, and shared-drive records.
  • Every AI answer must cite the source record and stay inside a private document set.
How we would start

A comparable engagement starts with a scoped ingestion of a real document set and a grounded-chat proof over it — enough to judge whether the answers are trustworthy on your own material before building the full workflow.

Start a conversation → Next Camping rental & trip planning →