Quickstart

Welcome to Inconvo, your go-to platform for enabling Ask-AI for your customers. In this example we will walk you through the essentials of connecting your data, managing tenancy, evaluating your Ask-AI and deploying it in your product.

Connect data:

We’ve already set up a sample database connection to help you quickly explore and test the platform without needing to connect your own data source.

If you would like to connect to a different data source, please see how to connect

Sample database

This database is a slight modification of metabase/qa-databases:postgres-sample-14 Metabase make the source code available here.

The Sample Database is a small database that ships with Inconvo so you can start asking questions before you connect to your database. The Sample Database contains four tables: Orders, People, Products, and Reviews.

Diagram

Orders

Confirmed Sample Company orders for a product, from a user.

CREATE TABLE orders (
    id int4 NOT NULL,
    user_id int4 NULL,
    product_id int4 NULL,
    subtotal float8 NULL,
    tax float8 NULL,
    total float8 NULL,
    discount float8 NULL,
    created_at timestamp NULL,
    quantity int4 NULL,
    CONSTRAINT orders_pkey PRIMARY KEY (id)
);

ALTER TABLE orders ADD CONSTRAINT fk_orders_product_id_products FOREIGN KEY (product_id) REFERENCES products(id);
ALTER TABLE orders ADD CONSTRAINT fk_orders_user_id_people FOREIGN KEY (user_id) REFERENCES people(id);

People

Information on the user accounts registered with Sample Company.

CREATE TABLE people (
    id int4 NOT NULL,
    address varchar(255) NULL,
    email varchar(255) NULL,
    "password" varchar(255) NULL,
    "name" varchar(255) NULL,
    city varchar(255) NULL,
    longitude float8 NULL,
    state bpchar(2) NULL,
    "source" varchar(255) NULL,
    birth_date date NULL,
    zip bpchar(5) NULL,
    latitude float8 NULL,
    created_at timestamp NULL,
    CONSTRAINT people_pkey PRIMARY KEY (id)
);

Products

Information on the user accounts registered with Sample Company.

CREATE TABLE products (
    id int4 NOT NULL,
    ean bpchar(13) NULL,
    title varchar(255) NULL,
    category varchar(255) NULL,
    vendor varchar(255) NULL,
    price float8 NULL,
    rating float8 NULL,
    created_at timestamp NULL,
    CONSTRAINT products_pkey PRIMARY KEY (id)
);

Reviews

Information on the user accounts registered with Sample Company.

CREATE TABLE reviews (
    id int4 NOT NULL,
    product_id int4 NULL,
    reviewer varchar(255) NULL,
    rating int2 NULL,
    body text NULL,
    created_at timestamp NULL,
    CONSTRAINT reviews_pkey PRIMARY KEY (id)
);

ALTER TABLE public.reviews ADD CONSTRAINT fk_reviews_product_id_products FOREIGN KEY (product_id) REFERENCES products(id);

Evalutate:

We provide evalauation as a first class citizen in the Inconvo platform.

We do this because to us accuracy is everything and we want to show you exactly how well Inconvo answer your users’ questions on your data.

Use the playground

Start by asking a question on the sample database within the platform playground.

Use the playground to ask any question of the connected database, like:

• “What are my best selling products?”
• “How many orders were made each month so far this year?”
• “What was the total discount given in Q4 2024?”
• What age group was the Synergistic Granite Chair most popular with in 2025?

Simply type your question in the input box and press enter.

Submit feeback

Upvote or downvote the response to provide feedback.

This feedback will be used to improve Inconvo over time.

Add to annotation queue

Simply click the plus button directly within the playground to add the response to the annotation queue.

Annotate examples

In the annotation queue you can see all of the examples you have added from the playground.

If the answer from the playground is incorrect, you can annotate the correct answer by clicking the edit button.

Once you have a correct answer, click the Add to Dataset button to add the example to your evaluations dataset.

View runs

Measure the performance of Inconvo on your data over time at a glance with evaluation runs.

Call the API:

Just need to send a POST request to our Ask API endpoint.

Request

curl --request POST \
    --url "https://api.inconvo.com/ask" \
    --header "Content-Type: application/json" \
    --data '{
      "question":
        "How many products have I sold per day of the week?",
      "conversationId": "conv_c9b1d",
      "requestContext": {
        "userId": "user_d7adc",
        "orgId": "org_99548"
      }
    }'

Response

{
    "type": "chart",
    "message":
        "Here's a breakdown of the number of
        products sold per day of the week.",
    "chart": {
      "type": "bar",
      "title": "Movies Rented Per Day of the Week",
      "xLabel": "Day of the Week",
      "yLabel": "Number of Movies Rented",
      "data": [
        {
          "label": "Monday",
          "value": 8
        },
        {
          "label": "Tuesday",
          "value": 8
        },
        {
          "label": "Wednesday",
          "value": 10
        },
        {
          "label": "Thursday",
          "value": 12
        },
        {
          "label": "Friday",
          "value": 10
        },
        {
          "label": "Saturday",
          "value": 10
        },
        {
          "label": "Sunday",
          "value": 6
        }
      ]
    }
}

Closing words

That’s it! There are only three steps you should really care about. Connecting, evaluating and calling the API.

All of them are covered here.