Smart Consultant
An AI consultant supporting customer communication on the web and WhatsApp. Data sources, channel connections and human handover are validated within the project scope.
Estate Pilot AI brings residential project information, customer communication and sales workflows together. The consultant is configured to use recorded sources for prices and unit data. It helps clarify buyer needs, prepare a lead record and hand the conversation to the sales team when needed.

An AI consultant supporting customer communication on the web and WhatsApp. Data sources, channel connections and human handover are validated within the project scope.
Agent links with serial numbers and first-click attribution. The aim is to record who introduced the buyer; recording conditions are checked during setup.
A matching layer where developers enter their projects and commission rates and agents see them in one place. Not built yet and not part of the current package.
Two separate leaks that need two separate answers.
A buyer's interest does not follow office hours. A question that arrives at midnight goes unanswered until morning, and by then the buyer has looked at another project.
Answering does not settle it either. Each sales representative quotes from memory: sometimes from an old list, sometimes rounded. If the price is set in euros, the lira equivalent drifts during the day and two representatives quote two different figures for the same unit.
On the agency side the problem is older. Who brought the buyer is often not recorded anywhere. That uncertainty produces commission disputes and keeps agents at arm's length from the project.
Estate Pilot AI addresses these two points separately. Information comes from one source, and who brought the buyer is recorded on the first click.
The consultant works in two places: the chat window on your website and your WhatsApp line.
It greets the buyer, asks what they are looking for and qualifies them: number of rooms, budget range, owner occupation or investment, and timing. It asks at most two questions per message and keeps sentences short.
Pricing responses are configured to use the recorded list. For euro-based lists, the lira equivalent uses a current rate and explains currency changes. The fallback to a euro-only price when a rate is unavailable is checked in acceptance scenarios.
Turkish is the default. Buyers who write in English or Russian are answered in their own language.
Once interest is clear it asks for a name and a phone number. Immediately before asking for the number it states the purpose of the record in one plain sentence, then opens the record and closes the conversation.


The consultant is configured to use project data returned by defined tools when composing a reply. Handover when information is missing or a tool fails is checked in acceptance scenarios.
Returns the official list price by unit type together with the lira equivalent at the day's live rate.
Queries the inventory by rooms, price range, floor and sales status.
Returns every field of a single unit by its number.
Name, location, concept and target handover date.
A tool retrieves render links registered to the project; the response is checked against its source image.
Name, phone, room preference, budget, purpose, timing and conversation notes are written into a single record.
On topics it cannot answer it moves the record to the human handover stage and writes down the reason.
These limits define intended behaviour. Data sources, tool responses and human handover are assessed through acceptance tests; results apply to the scope actually checked.
Price, area, floor and status are intended to come from the recorded database; values are compared with source records during checks.
It is configured to avoid discounts or campaign prices outside the recorded scope and to hand negotiation requests to a person.
It makes no commitments on financing, interest, title deeds or legal process.
It shares the handover date only as a recorded target, never as a guarantee.
It does not comment on or disparage competing projects.
It does not hide being an AI when asked.
When information is missing, the intended response is to explain the limit and hand the topic to a consultant rather than guess.
It never sends unsolicited promotional messages. It only replies to conversations the buyer starts.
Each agent is assigned a link with a serial number. A visit through that link creates an attribution record and a device trace. The configured ninety-day retention and browser access conditions are checked during setup.
The aim is to preserve a valid first attribution when later links arrive. Link sharing and serial-number behaviour are checked in acceptance scenarios.
Device sub-traces help track link use across different devices. A device record alone is not a count of individual people.
Serial numbers are checked against the records. Rejection of invalid traces and recovery of lost traces are checked in setup acceptance scenarios.
The system does not touch commission. It makes no payment, sets no rate and does not stand in the middle. It only documents who brought the buyer. The commission relationship stays entirely between the developer and the agent.
Estate Pilot AI is multi-tenant. More than one company's project lives in the same database. The critical question is this: under what circumstances can one company's data become visible to another.
The data boundary is designed to use PostgreSQL row level security alongside application controls. Restricted roles, company identity within the transaction and relevant table policies are assessed together.
The aim is to deny access when company identity or appropriate permission is missing. Incorrect identity and permission scenarios are tested; passing a test does not eliminate every possible fault or attack.
Temporary test companies are used to check access in both directions, unauthorized requests and cleanup of test data. Scenarios, versions and results are reviewed together. The test scope for a specific setup can be discussed during an evaluation call.
Decisions that answer the security and data protection questions before they are asked.
AWS Frankfurt is part of the product infrastructure approach. Actual data residency, backups and integration flows are verified for each setup.
For setups using Amazon Bedrock, the model, account and region settings are assessed together. Data flows and provider terms are checked within the project scope.
AWS Secrets Manager, database encryption and nightly backups are part of the infrastructure approach. Access permissions and restoration conditions are checked during setup.
Meta signature verification, a queue and a separate worker support the message flow. Duplicate-message handling and retry behaviour are tested.
UTC dates and cent-precision monetary fields are used. Currency conversions and rounding results are compared with source values in acceptance scenarios.
Why each resource exists is written in the repository, and the cost table is updated with every new resource. A cost alarm is in place from day one.
A measurable answer to the question of trusting an AI.
Acceptance scenarios are defined to cover pricing accuracy, tool failures, languages and lead flows. The tested setup is checked against the version intended for release.
Tested prices and area values are compared with the source database. Deviations are treated as errors to be addressed and retested.
Example scenarios include discount and final-price requests, maintenance fees, financing guarantees, competitor comparisons, English and Russian conversations, image requests and lead-record flows.
The lead flow checks whether the personal-data explanation appears before a phone number is requested. Findings are addressed and the relevant scenarios reassessed.
The commercial relationship is set out plainly.
Estate Pilot AI is used on a monthly software rental. Our revenue is the rental.
We take no share of sales. We take no share of commission. We do not enter the payment between you and the agent. The system records who brought the buyer; the rest is your commercial decision.
This choice is deliberate. A fee tied to sales volume makes the software a partner in the sale and creates a conflict of interest with both you and the agent. A fixed rental keeps the system neutral. The condition for trusting the attribution record is that the party keeping it takes no share of the outcome.
Infrastructure costs are transparent. What each AWS resource costs per month is laid out in a table, and what happens as usage grows is discussed in advance.
Three things are needed to start: your current price list, project information and the images you want to share.
The price list is loaded and configured as a data source for the consultant. Updates are tested through tool responses and the conversation.
Serial numbered links are generated for your agents and a chat window is added to your website. If the WhatsApp line is to be connected, Meta verification starts in parallel; that step can take a few days, so it begins on day one.
Behaviour scenarios, data isolation and legal texts are reviewed before release. Acceptance scope and responsibilities are agreed for the setup.

Said upfront, so nobody wastes time.
If your sales approach changes the price depending on the buyer, this does not fit. The consultant knows one list price and does not negotiate.
If you expect the AI to close the sale on its own, it does not fit. The consultant brings the buyer to the point of a call back; a human picks up the phone and a human signs the contract.
If inventory and pricing are not kept in order, the system will not fix that, only make it visible. Whatever the data says, the consultant says.
If you sell single units or resale property, the scale does not fit. The system was designed for selling many units within one project.
Reading from the database aims to reduce error risk; it is not a guarantee of error-free answers. The list, tool response and quoted price are checked together. Deviations are treated as errors, and the euro-only fallback when a rate is unavailable is also tested.
The consultant is configured to hand negotiation and out-of-scope discount requests to a person. Recording the reason and passing the conversation to the sales team are checked in acceptance scenarios.
PostgreSQL row level security, restricted roles and company identity checks are used together. Access in both directions and unauthorized requests are tested for each setup. Results apply to the scenarios and version actually tested.
AWS Frankfurt is part of the infrastructure approach. Actual data residency, backups, the model provider and other integrations are verified within the setup scope. Versions using Amazon Bedrock also require account and region checks.
No. The system never touches commission: it makes no payment, sets no rate and does not stand in the middle. It only records who brought the buyer. The commission relationship is entirely between you and the agent. Our revenue is the monthly software rental alone.
A current price list, project information and images are needed. Meta verification is planned separately if WhatsApp is requested. Timing depends on data preparation, integrations and the scope of behaviour, isolation and legal checks.
Share your project name, unit count and sales channels. We can review the data sources, consultant flow and setup acceptance tests using your own price list.