Assignment for Internship - Backend

Credit Approval System

In this assignment you will be working on creating a credit approval
system based on past data as well as future transactions, the goal of this
assignment is to assess the proficiency with the Python/Django stack,
using background tasks as well handing operations on Databases.

1. Setup and Initialization

a) Setup
• For this assignment please use Django 4+ with Django Rest
Framework.
• There is no requirement to make a frontend for the
application.
• You need to build appropriate data models for the
application
• The entire application and all its dependencies should be
dockerized.
• The application should use a Postgresql DB
b) Initiliazation
You are provided with a “customer_data.xlsx” which is a table
of the existing customers with the following attributes
• customer_id
• first_name
• last_name
• phone_number
• monthly_salary
• approved_limit
• current_debt
Also you are provided with “loan_data.xlsx” which is a table of
past and exisiting loans by customers with the following
attributes
 • customer id
• loan id
• loan amount
• tenure
• interest rate
• monthly repayment (emi)
• EMIs paid on time
• start date
• end date
Injest the provided data into the initial system using
background workers

2. API-
you need to build the following API endpoints with appropriate error
handling for and status codes for each.
Use compound interest scheme for calculation of monthly interest.
• /register
Add a new customer to the customer table with approved limit
based on salary using the following relation:

▪ approved_limit = 36 * monthly_salary (rounded to

nearest lakh)

a) Request body
Field Value
first_name First Name of customer
(string)
last_name Last Name of customer
(string)
age Age of customer (int)
monthly_income Monthly_income of individual
(int)
phone_number Phone number(int)

b) Response body
 Field Value
customer_id Id of customer (int)
name Name of customer (string)
age Age of customer (int)
monthly_income Monthly_income of individual
(int)
approved_limit Approved credit limit (int)
phone_number Phone number (int)

• /check-eligibility
Check loan eligibility based on credit score of customer (out of
100 ) based on the historical loan data from “loan_data.xlsx”,
consider the following components while assigning a credit
score:
i. Past Loans paid on time
ii. No of loans taken in past
iii. Loan activity in current year
iv. Loan approved volume
v. If sum of current loans of customer > approved limit of
customer , credit score = 0
Based on the credit score of the customer , approve loans as
per the following:
▪ If credit_rating > 50 , approve loan
▪ If 50 > credit_rating > 30 , approve loans with
interest rate > 12%
▪ If 30> credit_rating > 10 , approve loans with interest
rate >16%
▪ If 10> credit_rating , don’t approve any loans
▪ If sum of all current EMIs > 50% of monthly salary ,
don’t approve any loans
▪ If the interest rate does not match as per credit
limit , correct the interest rate in the response, i.e
suppose credit_limit is calculated to be 20 for a
 particular loan and the interest_rate is 8%, send a
corrected_interest_rate = 16% (lowest of slab) in
response

a) Request body
Field Value
customer_id Id of customer (int)
loan_amount Requested loan amount
(float)
interest_rate Interest rate on loan (float)
tenure Tenure of loan (int)

b) Response Body
Field Value
customer_id Id of customer (int)
approval can loan be approved (bool)
interest_rate Interest rate on loan (float)
corrected_interest_rate Corrected Interest Rate
based on credit rating , same
as interest rate if the interest
rate matches the slab (float)
tenure Tenure of loan (int)
monthly_installment Monthly installment to be
paid as repayment (float)

• /create-loan
Process a new loan based on eligibility.
a) Request body
Field Value
customer_id Id of customer (int)
loan_amount Requested loan amount
(float)
interest_rate Interest rate on loan (float)
tenure Tenure of loan (int)
 b) Response Body
Field Value
loan_id Id of approved loan, null
otherwise (int)
customer_id Id of customer (int)
loan_approved Is the loan approved (bool)
message Appropriate message if loan
is not approved (string)
monthly_installment Monthly installment to be
paid as repayment (float)

• /view-loan/loan_id
View loan details and customer details
a) Response Body
Field Value
loan_id Id of approved loan (int)
customer JSON containing id ,
first_name , last_name,
phone_number, age of
customer (JSON)
loan_amount Is the loan approved (bool)
interest_rate Interest rate of the approved
loan (float)
monthly_installment Monthly installment to be
paid as repayment (float)
tenure Tenure of loan (int)

• /view-loans/customer_id

View all current loan details by customer id

b) Response Body (list of loan items , each loan item will have
the following body)
 Field Value
loan_id Id of approved loan (int)
loan_amount Is the loan approved (bool)
interest_rate Interest rate of the approved
loan (float)
monthly_installment Monthly installment to be
paid as repayment (float)
repayments_left No of EMIs left (int)

3. General Guidelines
• Ensure code quality , organisation and segregation of responsibilities.
• Adding unit tests although is not nessecary, will be considered for bonus
points.
• The assignment should be submitted within 36 hours.
• The entire application and all its dependencies like DB should be
dockerized and should run from a single docker compose command.
• Please submit the github link of the repository.