Our goal is to give you a small coding challenge that gives you a chance to show off your skills while giving you an idea of some of the problems that you may encounter at Mindflash. We know you're busy with life, so we hope that you can spend around 2 hours working through this exercise. We don't expect you to finish in 2 hours, so don't worry if you can't. Submit what you have along with some notes on your thoughts and how you would proceed if you had more time. Most importantly, try to have some fun with it!

Lastly, your code is yours to keep, publish, delete, or blog about. However, please no mentions of Mindflash should you choose to share or publish it.

Your task is to build a small application that filters, transforms, and indexes a stream of events from NSQ (a real-time distributed messaging platform) into Elasticsearch (a highly performant and scalable search/aggregation engine).

This repository includes a number of resources to help you get started, including documentation, code samples, and a docker compose file that you can use to set up your local/test environment.

You just got hired to join the badass engineering team at BrainSpark (a fictional LMS)! The first story in your sprint backlog is to build an application that indexes a stream of events into ElasticSearch for use in reporting and anaytics. For reference, the domain model of our application for the exercise is shown below:

At a high level:

• Our product supports multiple accounts
• Each account has many courses and many users
• A user may attempt a course many times
• Users can perform actions that trigger events

Our focus for this exercise is the domain events. Events are published on the events NSQ Topic. An event message represents a single event and is serialized as JSON. A sample event is shown below:

{
    "id": "5c92de28-14f0-449e-821b-e61e871179c2",
    "attempt_id": "e70a6ecf-f308-4306-831d-bb41c851061d",
    "progress": 0.63414633,
    "score": 0.96158886,
    "timestamp": "2018-01-16T14:09:51.655185082-07:00",
    "type": "PROGRESS",
    "user_id": "471de972-520c-4f44-bd6d-34cc98dd5e6e"
}

Prior to indexing the event records into ES, you will need to apply the following transformations:

• filter out invalid messages (ie messages that do not pass the schema above)
• filter out events with type FOO
• denormalize the event by hydrating (ie embedding) the related attempt, course, trainee, and user (if available) using the BrainSpark JSON-RPC API (see resources section below for more info)

An example of a successful transformation is shown below:

{
    "id": "5c92de28-14f0-449e-821b-e61e871179c2",
    "timestamp": "2018-01-16T14:09:51.655185082-07:00",
    "type": "PROGRESS",
    "attempt_id": "e70a6ecf-f308-4306-831d-bb41c851061d",
    "progress": 0.63414633,
    "score": 0.96158886,
    "user_id": "471de972-520c-4f44-bd6d-34cc98dd5e6e",
    "attempt": {
        "id": "e70a6ecf-f308-4306-831d-bb41c851061d",
        "course_id": "9a3c3136-c24d-4fb6-94fe-7ba6aa491080",
        "created_at": "2017-07-09T00:45:24.466Z",
        "trainee_id": "747f457c-7e74-400e-9b0e-071d1c693ea4",
    },
    "course_id": "9a3c3136-c24d-4fb6-94fe-7ba6aa491080",
    "course": {
        "id": "9a3c3136-c24d-4fb6-94fe-7ba6aa491080",
        "account_id": "aacf3f09-4b19-424f-995a-4b13f8e36365",
        "name": "hot dog vs not hot dog"
    },
    "trainee_id": "747f457c-7e74-400e-9b0e-071d1c693ea4",
    "trainee": {
        "id": "747f457c-7e74-400e-9b0e-071d1c693ea4",
        "account_id": "aacf3f09-4b19-424f-995a-4b13f8e36365",
        "first_name": "erlich",
        "last_name": "bachman",
        "email": "[email protected]"
    },
    "user": {
        "id": "471de972-520c-4f44-bd6d-34cc98dd5e6e",
        "account_id": "aacf3f09-4b19-424f-995a-4b13f8e36365",
        "first_name": "jin",
        "last_name": "yang",
        "email": "[email protected]"
    }
}

The events need to be indexed into Elasticsearch using event as the document type and index events-YYYY-MM-DD, where YYYY-MM-DD is replaced using the formatted event timestamp attribute. Note: you do not need to preallocate the indices, they will be created on first index.

• NSQ Design Overview
  • nsqjs node client
  • go-nsq go client
• Elasticsearch Document API
  • elasticsearch.js node client
  • elastic go client
• BrainSpark API
• JSON RPC
• a sample json event

The only prerequisites for using this repository are Docker & Compose. Installation links are below:

• Docker
• Compose

1. Clone this repository
2. Write your code
3. Submit

To start the local environment:

$ docker-compose up -d

Note: This command will start NSQ, ElasticSearch, and the BrainSpark container in the background. It may take a few minutes the first time you run it, as it will have to pull the images for NSQ and ElasticSearch and build the BrainSpark image locally. By default, NSQ will be mapped to localhost:4150, ElasticSearch will be mapped to localhost:9200, and the BrainSpark application will be listening on port 8080. You can change these mappings and other configurations by editing the docker-compose.yml file

Caution: If you're not actively consuming messages, NSQ will continue to queue messages, first in memory, and then on disk. You can tweak the speed and volume of messages to NSQ by modifying the brainspark config. You can monitor NSQ by visiting nsqadmin in your browser. You can also control the message flow by starting and stopping the brainspark container.

To stop all containers:

$ docker-compose down

To run start/stop individual containers:

$ docker-compose stop brainspark
$ docker-compose start brainspark

• your application should be writtin in node.js or go
• you should include a Dockerfile in the root of your repository
• your code should be linted
  • nodejs: use eslint with eslint-config-airbnb-base
  • go: use golint
• your code should include at least a couple of tests
• your code should include a README.md file in the root with instructions for building, running, and testing. It can also include notes on your throught process and any issues you may have run into.

We will evaluate your submission using the following criteria

• Is your application well organized?
• Is your code documented?
• Is your code efficient & performant?

Please upload this repository to Github and submit to @jgiless when complete. Also, we would love your feedback, so feel free to share your thoughts on the exercise!