Agentic Listing and Proposal Portal for Property Sales
Salespeople drop in brochures and photos or simply talk, and the portal drafts the listing, checks the photos and builds the client proposal.
A listing and proposal portal for property salespeople in Singapore covering homes, commercial and industrial space. It reads fact sheets and photographs together, fills the listing with a source for every value, and asks only for the mandatory fields still missing. Photos pass quality checks with plain reasons when rejected. A client brief, typed or spoken, returns ranked matches with the reason each one fits, and the proposal, letter of intent and option to purchase are built from fixed templates.
Salespeople assemble brochures, photos and specifications by hand and send them over messaging apps. Listings arrive incomplete, photo quality varies, and every client proposal is built from scratch. The brief asked for a portal that would cut the time to list and the time to propose without loosening compliance with Singapore property rules.
A voice assistant sits on every page. Extraction reads PDFs, Word files and photos at once and shows where each value came from. Photo checks reject low-resolution or blurred images and suggest a fix. Matching filters on hard requirements and ranks on preferences. Stamp duties, eligibility and document clauses come from deterministic rules, and instructions planted inside an uploaded fact sheet are ignored.
A complete sample listing was drafted from its files in 1.4 minutes against a target of 8, and a client brief became a proposal in 2.1 minutes against a target of 3. Both figures come from one sample listing and two proposals, so they show the approach rather than a production result. The planted instruction in the sample fact sheet was ignored.
- Status
- Working system
- Client
- Confidential
Built to the Governed Cognitive Layer: the model reads and drafts, deterministic checks decide what it may do, and a named person approves anything that commits the business.
Point AI at the slow, repetitive part of the job and keep the regulated part on rails. Here the model reads documents and drafts text, while stamp duty, eligibility and contract clauses come from fixed rules and templates. Show the source of every value so the salesperson can check it in seconds, and treat anything inside an uploaded document as data, never as an instruction.
Concept paper, C-suite deck and live demo
The concept paper, the deck and the working portal are available on request. The walkthrough drafts a listing from its files live.
●Live demo (on request) →The same design rules behind this build, applied to your business in half a day.
Book the WorkshopDiscuss a build like thisScreens from the system
Demo data throughout. Customer names, accounts and payees are fictional.
Other projects
Agentic AIAgentic Auction Portal for Certified Coins
Collectors bid on certified coins online while AI agents check certificates, describe lots and flag suspicious bidding, and staff release every payout.
See how it works →
Agentic AITandem: Voice Practice for Children Between Therapy Sessions
Children practise the speech goals their therapist has set, at home between sessions, with a gentle voice character, a parent beside them and no audio kept.
See how it works →
Agentic AIOneness: A Private AI Companion with Safety Built Around It
Adults talk privately about work, relationships and decisions, by text or voice, with every message checked for risk before any reply and a human listener one consent away.
See how it works →
In testing, a brief asking for a home available by March came back searching for March 1970. The server had fixed today's date once at start-up, before any request arrived, and read it as the start of computer time. It is a small bug, but a proposal with the wrong date in front of a buyer is a credibility problem. Dates now come from each request.
