Write your pricing, eligibility, or fraud policy in plain English. SmartDecision turns it into a decision table your team can read and edit, and your apps call over an API in milliseconds. It runs on your own servers with a local LLM, so nothing leaves your network.
Founding-member pricing and a direct line to the team. One email when it launches, no spam.
One loan policy, three forms: the sentence your team writes, the table it becomes, and the call your app makes.
“A loan is approved if credit score is 700 or higher, age is between 21 and 65, and income is at least €10,000. Scores from 650 to 699 go to manual review.”
| credit_score | age | income | decision |
|---|---|---|---|
| >= 700 | 21...65 | >= 10000 | approved |
| 650...699 | 21...65 | >= 10000 | review |
| * | * | * | rejected |
POST /api/resolve/214
credit_score = 712
age = 34
income = 18000{ "decision": "approved" }
Pricing tiers, eligibility checks, fraud thresholds, approval limits. Each one starts as a simple
if statement and ends as nested logic nobody wants to touch. The people who own the policy
can't read it, and every change waits in the engineering queue.
SmartDecision moves those decisions out of the codebase and into tables that business teams can read and edit, engineers can test, and applications can call.
Describe the rule the way you'd explain it to a colleague. The local LLM drafts a decision table with inputs, outputs, and rules. You review it and change any cell by hand.
New decision
Decision policy
A loan is approved if credit score is 700 or higher, age is between 21 and 65, and income is at least €10,000. Scores from 650 to 699 go to manual review.
The prompt should be a clear and comprehensive statement of the decision policy.
Save the outcomes you expect as tests, like “a score of 700 is approved”. Run them before every change, so adjusting one rule never quietly breaks another.
| Test input | Expected output | Updated | Actions |
|---|---|---|---|
| credit_score = 700 | approved | now | |
| credit_score = 640 | rejected | now |
Your applications send inputs to one REST endpoint with an API key and get the answer back in milliseconds. Answers come from plain rule-matching: the LLM drafts tables, it never answers requests.
Resolve API
const formData = new FormData();
formData.append("credit_score", "712");
fetch("https://bi.smartdecision.io/api/resolve/214", {
method: "POST",
headers: {
Authorization: "HV3g…LqU5",
},
body: formData,
});Important: the resolve API key is a secret that should be kept confidential. Share it only with trusted parties.
The questions your engineers will ask, answered up front.
cp.POST /api/resolve/:id with an API key. Works from any language that can make an HTTP request.Discounts, promotions, regional prices, tier thresholds. Changed by the people who set them, not by a deploy.
Who can use a feature, upgrade a plan, or apply for a service. Defined once, answered the same in every app.
Transaction thresholds and risk policies evaluated inside your network, so sensitive data stays put.
Gate features by user attributes, region, or experiment without redeploying the application.
Everything runs inside your infrastructure: the rule engine, the web console, and the LLM that drafts your rules. Credit scores, transaction amounts, eligibility inputs: none of it crosses your network boundary, not when you write a rule and not when an app resolves one. When compliance asks “where does the data go?”, the answer is nowhere.
SmartDecision isn't on sale yet. Join the waitlist to hear the moment it is, and tell us what your team needs before we ship.
Founding-member pricing and a direct line to the team. One email when it launches, no spam.