Let's imagine a (almost) hypothetical situation where:
- lots of consultants are waiting to be attributed to a mission
- business managers are flooded with mission requests
- once a business manager decide to attribute a mission to a consultant, the consultant begin its mission immediatly (no client inteview, no time to prepare contract, ...)
Given a list of mission and consultants, the goal is to maximize CBTW's revenue. The consultants are represented by a list of skills. The missions are repesented by a nominal daily rate and a list of wanted skills. For each mission, the earned money is the pourcentage of wanted skills the consultant possess multiply by the nominal daily rate.
For example, if there is only 1 mission with 5 required skills and only 1 consultant with 4 of those skills, the client will pay 80% of the nominal daily rate.
The program takes a JSON as input (from stdin) and output a JSON (to stdout)
The input looks like:
{
"consultants": [
{
"id": "0",
"name": "Andrew Williams",
"skills": [
"python",
"java",
"rust"
]
},
{
"id": "1",
"name": "Robert Graham DVM",
"skills": [
"Git"
]
},
{
"id": "2",
"name": "Jason Russell",
"skills": [
"java",
"rust",
"Git"
]
}
],
"missions": [
{
"id": "0",
"client": "Sanchez-Lopez",
"skills": [
"French",
"TDD",
"python",
],
"rate": 276
},
{
"id": "1",
"client": "Spencer and Sons",
"skills": [
"Dutch",
"SQL",
"rust"
],
"rate": 552
},
]
}
The output contains a mapping {consultant_id: mission_id}
{
"0": "0",
"2": "4"
}
Boiler plate is provided for python.
Complete the file python/cbtw.py
.
If you use another language, contact me
and I'll add the corresponding boiler plate.
Github actions run tests for you when you push your code.
See the github actions to know how your code performed.
Several steps have to be achieved:
- step 1: have a valid solution, i.e. make money.
- step 2: make as much money as possible.