Attempt 1 outlined in blog post: https://www.tonic.ai/post/how-to-generate-simple-test-data-with-faker/

attempt1.py

import random
from datetime import timedelta, datetime
from faker import Faker
from faker.providers import person
from faker.providers import internet
from faker.providers import ssn
from faker.providers import address
from faker.providers import job
from faker.providers import date_time

fake = Faker()
fake.add_provider(person)
fake.add_provider(internet)
fake.add_provider(ssn)
fake.add_provider(address)
fake.add_provider(job)
fake.add_provider(date_time)
numRows = 10

d = dict()
d['first_name'] = lambda: fake.first_name()
d['last_name'] = lambda: fake.last_name()
d['gender'] = lambda: 'M' if random.randint(0,1) == 0 else 'F'
d['personal_email'] = lambda: fake.email()
d['ssn'] = lambda: fake.ssn()
#Note that on windows you cannot currently use the date_time functionality in Faker due to a known bug.  Comment out the following two lines and uncomment the two lines below that.
#d['birth_date'] = lambda: fake.date_between_dates(date_start=datetime(1960, 1, 1), date_end=datetime(2000, 1, 1))
#d['start_date'] = lambda: fake.date_between_dates(date_start=datetime(1995, 1, 1), date_end=datetime(2019, 1, 1))
d['birth_date'] = lambda: (datetime(1960, 1, 1) + timedelta(seconds=1261600000)).strftime('%m/%d/%Y')
d['start_date'] = lambda: (datetime(1995, 1, 1) + timedelta(seconds=725420000)).strftime('%m/%d/%Y')
d['office'] = lambda: fake.city()
d['title'] = lambda: fake.job()
d['org'] = lambda: random.choice(['Engineer','Sales','Associate','Manager','VP'])
d['accrued_holidays'] = lambda: random.randint(0,20)
d['salary'] = lambda: round(random.randint(90000,120000)/1000)*1000
d['bonus'] = lambda: round(random.randint(0,5000)/500)*500

for _ in range(numRows):
    print([d[k]() for k in d.keys()])