Представьте себе, что вы пытаетесь устроиться на первую работу
Java Junior Dev в одну из компаний.
Вы прошли первое интервью с HR специалистом и после интервью
получили реальное домашнее задание для проверки ваших технических
навыков.
Обычно технические специалисты ИТ компаний очень заняты решением
реальных бизнес задач, поэтому у них нет времени на проведение
собеседований со всеми кандидатами на работу. А кандидатов как
вы знаете действительно много! Но кандидаты по уровню технических
навыков бывают разные. Компании хотят общаться с лучшими!
Для отсеивания/отбора кандидатов компания выдаёт техническое задание.
Кандидат решает задание и присылает решение на проверку. Технический
специалист компании оценивает качество решения и принимает решение
о проведении или не проведении технического интервью с кандидатом.
Ваша цель: решить техническое задание так, чтобы доказать
работодателю, что вас имеет смысл пригласить на техническое
интервью.
В процессе работы над этим заданием особое внимание будет уделено:
- этапам процесса разработки;
- бизнес анализу задачи;
- моделированию данных;
- реализации алгоритма;
- TDD (Test Driven Development);
- качеству кода (clean code);
- качеству архитектуры;
- удобству поддержки и расширяемости и решения.
Ваша первая задача: ознакомьтесь с описанием задачи:
Introduction
This task is created with intention to test your Java coding and analytical skills when you're at home, without stress and without any distractions.
Do this task at your own pace.
Please read task description carefully and implement needed functionality.
Write code by keeping in mind the best code writing and testing practices (we value clean code, clean architecture, good code readability very much).
Make sure your code is tested and working according to task description.
Write unit tests.
Things you may use to achieve the needed result:
Java 8+
jUnit 4+
Framework or library that supports dependency injection
Any source on the internet that may help you
Expected result:
Java 8+ project
Solution hosted on GitHub (preferred) or compressed as zip file
Unit tests for the implementation
Please use build tool such as Maven or Gradle
Implementation description
Task description:
Consider this description as a business task description in issue tracking system (e.g. Jira). All the analysis was done by system analysts and following description was created.
Insurance company wants to start issuing private property policies to their customers. System analysts found out that there will be a policy which will have objects (e.g. a House) and that objects will have sub-objects (e.g. electronic devices such as TV). One policy can contain multiple objects. One object can contain multiple sub-objects. In this iteration, customer needs a functionality that calculates premium (a price that will be paid by each client that will buy this insurance) for the policy.
Premium is calculated by a formula defined in "Needed functionality" section.
In short - formula groups all sub-objects by their type, sums their sum-insured and applies coefficient to the sum. Then all group sums are summed up which gets us a premium that must be paid by the client.
No GUI is needed, policy data will be sent through the API directly to the methods that will be created. No database is needed, functionality should not store any data. It should receive policy object, calculate premium and return result. Preferred invocation of the functionality but may be changed if needed:
PremiumCalculator#calculate(Policy policy);
Needed functionality:
Please create functionality that calculates policy premium.
In this iteration client stated that only risk types FIRE and THEFT will be calculated, however it may be possible that in near future more risk types will be added.
Make sure that it is easy to extend implementation for new risk types.
Premium calculation formula:
PREMIUM = PREMIUM_FIRE + PREMIUM_THEFT
PREMIUM_FIRE = SUM_INSURED_FIRE * COEFFICIENT_FIRE
SUM_INSURED_FIRE - total sum insured of all policy's sub-objects with type "Fire"
COEFFICIENT_FIRE - by default 0.014 but if SUM_INSURED_FIRE is over 100 then 0.024
PREMIUM_THEFT = SUM_INSURED_THEFT * COEFFICIENT_THEFT
SUM_INSURED_THEFT - total sum insured of all policy's sub-objects with type "Theft"
COEFFICIENT_THEFT - by default 0.11 but if SUM_INSURED_THEFT equal or greater than 15 then 0.05
Acceptance criteria:
If there is one policy, one object and two sub-objects as described below, then calculator should return
premium = 2.28 EUR
Risk type = FIRE, Sum insured = 100.00
Risk type = THEFT, Sum insured = 8.00
If policy's total sum insured for risk type FIRE and total sum insured for risk type THEFT are as described below, then calculator should return premium = 17.13 EUR
Risk type = FIRE, Sum insured = 500.00
Risk type = THEFT, Sum insured = 102.51
Object entities:
Policy:
Policy can have multiple policy objects and each policy object can have multiple sub-objects. Policy has 3 attributes:
- Policy number e.g. LV20-02-100000-5
- Policy status e.g. REGISTERED, APPROVED
- Policy objects - Collection of one or multiple objects.
Policy object:
Policy objects can have multiple sub-objects and can be related only to one policy
Policy objects have 2 attributes:
- Object name e.g. A House;
- Sub-objects Collection of none or multiple sub-objects.
Policy sub-object:
Policy sub-objects can be related only to one policy object and have 3 attributes:
- Sub-object name e.g. TV;
- Sum insured Cost that will be covered by insurance;
- Risk type e.g. FIRE, THEFT.
Для удобства работы над проектом мы создали GitHub репозиторий:
В этом репозитории вы найдёте поэтапное решение задачи (приведённой выше) состояшее из 26 шагов.
Если у вас появилось желание попробовать порешать поставленную задачу, вы можете присоединиться к нашей Телеграм группе (вот линк: https://t.me/+2E9gjF-BFsFjYWE0), которую мы создали как раз для этого небольшого проета. Там можно задать любые вопросы по проекту, попросить доступ к GitHub репозиторию на запись для полноценной разработки!