As part of the recruitment process for a backend developer position, you are required to complete a coding challenge to assess your skills in our tech stack, microservices in a monorepo including relational database, REST with Node.js and Express, GraphQL. The challenge consists of three tasks focusing on different aspects of backend development.
Note: Feel free to use any additional libraries or tools you are comfortable with.
The goal is to build API for a questionaire so that:
- It is customizable though a database
- Has GraphQL API that for a provided user input (answer) can give a recommendation of products
- Here is description of the logic we require can be found furthe down the challenge
Model the database entities by starting from a sample payload that looks like this:
{
selectorID: '43F54E00-B8F9-11EC-AB32-E792EE247845',
pages: [
{
pageID: '95086460-B8FB-11EC-AB32-E792EE247845',
urlEndpoint: '0',
conditionals: [],
questions: [
{
questionID: '95086461-B8FB-11EC-AB32-E792EE247845',
type: 'cards',
answers: [
{
answerID: '07152B60-B8FC-11EC-AB32-E792EE247845',
},
{
answerID: '9507D2B0-B8FC-11EC-AB32-E792EE247845',
},
],
},
],
},
{
pageID: '1EEB3350-B8FD-11EC-AB32-E792EE247845',
urlEndpoint: '1',
questions: [
{
questionID: '1EEB3351-B8FD-11EC-AB32-E792EE247845',
type: 'cards',
answers: [
{
answerID: '58402C00-B8FD-11EC-AB32-E792EE247845',
},
{
answerID: '04031E80-B8FE-11EC-AB32-E792EE247845',
},
{
answerID: '2A772CA0-B8FE-11EC-AB32-E792EE247845',
},
],
},
],
conditionals: [],
},
{
pageID: '688C6C80-B8FE-11EC-AB32-E792EE247845',
urlEndpoint: '2',
questions: [
{
questionID: '688C6C81-B8FE-11EC-AB32-E792EE247845',
type: 'buttons',
answers: [
{
answerID: '93F28030-B8FE-11EC-AB32-E792EE247845',
},
{
answerID: '95C36FA0-B8FE-11EC-AB32-E792EE247845',
},
],
},
],
conditionals: [],
},
{
pageID: 'A11A1D90-B8FE-11EC-AB32-E792EE247845',
urlEndpoint: '3',
questions: [
{
questionID: 'A11A1D91-B8FE-11EC-AB32-E792EE247845',
type: 'slider',
answers: [
{
answerID: '0C22AB60-B900-11EC-AB32-E792EE247845',
},
{
answerID: '1A242E00-B900-11EC-AB32-E792EE247845',
},
{
answerID: '2389BA50-B900-11EC-AB32-E792EE247845',
},
{
answerID: '2A2DB5A0-B900-11EC-AB32-E792EE247845',
},
],
},
],
conditionals: ['93F28030-B8FE-11EC-AB32-E792EE247845'],
},
{
pageID: '448AAC00-B900-11EC-AB32-E792EE247845',
urlEndpoint: '4',
questions: [
{
questionID: '448AAC01-B900-11EC-AB32-E792EE247845',
type: 'cards',
answers: [
{
answerID: 'B1F69AB0-B900-11EC-AB32-E792EE247845',
},
{
answerID: 'B8441190-B900-11EC-AB32-E792EE247845',
},
],
},
],
conditionals: ['95C36FA0-B8FE-11EC-AB32-E792EE247845'],
},
{
pageID: 'F18449C0-B900-11EC-AB32-E792EE247845',
urlEndpoint: '5',
questions: [
{
questionID: 'F18449C1-B900-11EC-AB32-E792EE247845',
type: 'cards',
answers: [
{
answerID: '19CCC9C0-B901-11EC-AB32-E792EE247845',
},
{
answerID: '41731650-B901-11EC-AB32-E792EE247845',
},
{
answerID: '54F7BAF0-B901-11EC-AB32-E792EE247845',
},
{
answerID: '6A22BC90-B901-11EC-AB32-E792EE247845',
},
{
answerID: '79A93180-B901-11EC-AB32-E792EE247845',
},
{
answerID: '8B8E56F0-B901-11EC-AB32-E792EE247845',
},
],
},
],
conditionals: [],
},
],
}
that is used by the frontend to visualize a questionnaire.
- The final database should have relational entities such as:
page,input,answer,product,result. - There should be more entities that allow expressing the question/answer types (see in the json that checkbox and button inputs have different attributes)
Using the ORM, provide initial migration and initial seed of the whole questionaire and the products.
Create a service layer that implements the business logic. The required microservices are:
questionnaire- responsible for CRUD operations on the questionnaire definition (page, input, answer)products- dataset of products for a questionaire, with API that has also support for pagination (For an example dataset: https://www.foxbase.de/en/getting-started-dokumentation/import-product-data/ <-- at the bottom of the page. Otherwise you can also use a public api for random products)user-input- responsible for capturing the sessions, requests and inputsrecommendations- where the logic of getting from answer to result is implemented. It should support resuming a questionaire (Here you can find a description: https://www.foxbase.de/en/getting-started-dokumentation/matrix-empfehlungslogik/)
It is not mentioned above, but assumed that:
- there can be more than one questionary in the system
- more than one user can be on the same questionary
- the results page should be shareable
Create a GraphQL API
- Recommendation API for guiding through the questionary
- Admin API to manage the questionary
- Support API to query statistics, with at least one query: completion rate per day (by reaching results page)
- Code must be written in TypeScript.
- Code is documented
- Application can build and run
- Tests
- Clean Code in your application (Example Guideline)
Please provide your solution as a fork of the repositoriy.
Matrix Recommendation Explanation
| Product A | Product B | Product C | Product D | |
|---|---|---|---|---|
| Answer A | 1 | 0 | 1 | 1 |
| Answer B | 0 | 1 | 0 | 1 |
| Answer C | 1 | 0 | 1 | 0 |
Answer A: given
Answer B: not given
Answer C: given
Product A: 1 * 1 = 1 <-- Recommend
Product B: 0 * 0 = 0 <-- Not Recommend
Product C: 1 * 1 = 1 <-- Recommend
Product D: 1 * 0 = 0 <-- Not Recommend
In essence, it's a Matrix x Vector multiplication
Example Input Request from the Frontend
{
"95086461-B8FB-11EC-AB32-E792EE247845": {
"07152B60-B8FC-11EC-AB32-E792EE247845": {
"answerID": "07152B60-B8FC-11EC-AB32-E792EE247845",
"creationTimestamp": 1691066231.123,
"modifiedTimestamp": 1691066231.123,
"value": null,
"skipped": false
},
"9507D2B0-B8FC-11EC-AB32-E792EE247845": {
"answerID": "9507D2B0-B8FC-11EC-AB32-E792EE247845",
"creationTimestamp": 1691066244,
"modifiedTimestamp": 1691066312.614,
"value": true
}
},
"1EEB3351-B8FD-11EC-AB32-E792EE247845": {
"58402C00-B8FD-11EC-AB32-E792EE247845": {
"answerID": "58402C00-B8FD-11EC-AB32-E792EE247845",
"creationTimestamp": 1691066236.586,
"modifiedTimestamp": 1691066314.5,
"value": 1,
"skipped": false
},
"04031E80-B8FE-11EC-AB32-E792EE247845": {
"answerID": "04031E80-B8FE-11EC-AB32-E792EE247845",
"creationTimestamp": 1691066244,
"modifiedTimestamp": 1691066244,
"value": null,
"skipped": false
},
"2A772CA0-B8FE-11EC-AB32-E792EE247845": {
"answerID": "2A772CA0-B8FE-11EC-AB32-E792EE247845",
"creationTimestamp": 1691066244,
"modifiedTimestamp": 1691066244,
"value": null,
"skipped": false
}
},
"688C6C81-B8FE-11EC-AB32-E792EE247845": {
"93F28030-B8FE-11EC-AB32-E792EE247845": {
"answerID": "93F28030-B8FE-11EC-AB32-E792EE247845",
"creationTimestamp": 1691066244,
"modifiedTimestamp": 1691066316.116,
"value": 'String',
"skipped": false
},
"95C36FA0-B8FE-11EC-AB32-E792EE247845": {
"answerID": "95C36FA0-B8FE-11EC-AB32-E792EE247845",
"creationTimestamp": 1691066239.331,
"modifiedTimestamp": 1691066239.331,
"value": null,
"skipped": false
}
}
Tree Structure Logic
The base of each questionnaire is the question or decision tree. A question tree consists of several question levels. These later lead the customer to the right product recommendation.
The logic of a question tree looks like this:
Tipps
- Please focus on sound architecture and a clean separation of concerns regarding your code ( do not put everything into one file)
- If you have any questions while doing the challenge create an issue in your repo where we can close the issue as soon I have given a sufficient answer to your question
- Product Data can be either gotten from here: https://fakestoreapi.com/ or similar free Apis