Jobhunter Simulation Script

jobhunter_sim.py

# This isn't shared as an example of beautiful code or anything!
# It was a quick and dirty script I wrote to simulate how the 
# job market might work for bootcamp grads given a few different
# assumptions.

# I'm sharing it so that you can open it up and plug in your own
# beliefs, either by changing the numbers or even adding new logic.
# Through this I hope you might test your own beliefs and their
# outcomes.

from dataclasses import dataclass
import random


@dataclass
class Candidate:
  period: int
  strength: float
  original_strength: float

  def __init__(self, period, strength):
    self.period = period
    self.strength = strength
    self.original_strength = strength

def report(hired, given_up, remaining):
  all_candidates = hired + given_up + remaining
  # Go through the strength in intervals of 0.05
  for i in range(20, 0, -1):
    start = (i - 1) * 5
    end = i * 5
    total_candidates_in_interval = len([c for c in all_candidates if start <= c.original_strength < end])
    if total_candidates_in_interval == 0:
      continue
    hired_pct = len([c for c in hired if start <= c.original_strength < end]) / total_candidates_in_interval * 100
    given_up_pct = len([c for c in given_up if start <= c.original_strength < end]) / total_candidates_in_interval * 100
    remaining_pct = len([c for c in remaining if start <= c.original_strength < end]) / total_candidates_in_interval * 100
    print(f"Strength: {start}-{end}, Hired: {hired_pct:.2f}%, Given up: {given_up_pct:.2f}%, Remaining: {remaining_pct:.2f}%")

# This is the first graph I show in the video
def simulate_constrained_hires():
  MONTHS_TO_SIMULATE = 120
  CHANCE_OF_GIVING_UP_EACH_MONTH = 0.1
  HIRES_MADE_PER_MONTH = 5
  STUDENTS_GRADUATING_PER_MONTH = 25

  jobhunters = []
  all_hired = []
  all_given_up = []
  for month in range(MONTHS_TO_SIMULATE):
    # Sort by strength
    jobhunters.sort(key=lambda c: c.strength, reverse=True)

    # Simulate hiring
    hired = jobhunters[:HIRES_MADE_PER_MONTH]
    all_hired += hired
    jobhunters = jobhunters[HIRES_MADE_PER_MONTH:]

    # Simulate giving up
    given_up = []
    for candidate in jobhunters:
      if random.uniform(0, 1) < CHANCE_OF_GIVING_UP_EACH_MONTH:
        given_up.append(candidate)
    jobhunters = [candidate for candidate in jobhunters if candidate not in given_up]
    all_given_up += given_up

    # Simulate graduation
    jobhunters += [Candidate(month, random.uniform(0, 100)) for _ in range(STUDENTS_GRADUATING_PER_MONTH)]

    # Print results
    print(f"Month {month} -> Graduated: {STUDENTS_GRADUATING_PER_MONTH}, Hired: {len(hired)}, Given up: {len(given_up)}, Remaining: {len(jobhunters)}")
  return all_hired, all_given_up, jobhunters

# This is the last graph I show in the video
def simulate_constrained_hires_with_improvement():
  MONTHS_TO_SIMULATE = 120
  CHANCE_OF_GIVING_UP_EACH_MONTH = 0.1
  HIRES_MADE_PER_MONTH = 5
  STUDENTS_GRADUATING_PER_MONTH = 25

  jobhunters = []
  all_hired = []
  all_given_up = []
  for month in range(MONTHS_TO_SIMULATE):
    # Sort by strength
    jobhunters.sort(key=lambda c: c.strength, reverse=True)

    # Simulate hiring
    hired = jobhunters[:HIRES_MADE_PER_MONTH]
    all_hired += hired
    jobhunters = jobhunters[HIRES_MADE_PER_MONTH:]

    # Simulate giving up
    given_up = []
    for candidate in jobhunters:
      if random.uniform(0, 1) < CHANCE_OF_GIVING_UP_EACH_MONTH:
        given_up.append(candidate)
    jobhunters = [candidate for candidate in jobhunters if candidate not in given_up]
    all_given_up += given_up

    # Simulate improvement
    for c in jobhunters:
      c.strength += 5

    # Simulate graduation
    jobhunters += [Candidate(month, random.uniform(0, 100)) for _ in range(STUDENTS_GRADUATING_PER_MONTH)]

    # Print results
    print(f"Month {month} -> Graduated: {STUDENTS_GRADUATING_PER_MONTH}, Hired: {len(hired)}, Given up: {len(given_up)}, Remaining: {len(jobhunters)}")
  return all_hired, all_given_up, jobhunters

# This is the middle graph I show in the video
def simulate_unconstrained_hires():
  MONTHS_TO_SIMULATE = 120
  CHANCE_OF_GIVING_UP_EACH_MONTH = 0.1
  STUDENTS_GRADUATING_PER_MONTH = 25
  CHANCE_OF_INTERVIEW = 0.25

  jobhunters = []
  all_hired = []
  all_given_up = []
  for month in range(MONTHS_TO_SIMULATE):
    # Sort by strength
    jobhunters.sort(key=lambda c: c.strength, reverse=True)

    # Simulate hiring
    for candidate in jobhunters:
      if random.uniform(0, 1) > CHANCE_OF_INTERVIEW and candidate.strength > random.uniform(20, 120):
        all_hired.append(candidate)
    jobhunters = [candidate for candidate in jobhunters if candidate not in all_hired]

    # Simulate giving up
    given_up = []
    for candidate in jobhunters:
      if random.uniform(0, 1) < CHANCE_OF_GIVING_UP_EACH_MONTH:
        given_up.append(candidate)
    jobhunters = [candidate for candidate in jobhunters if candidate not in given_up]
    all_given_up += given_up

    # Simulate graduation
    jobhunters += [Candidate(month, random.uniform(0, 100)) for _ in range(STUDENTS_GRADUATING_PER_MONTH)]

    # Print results
    print(f"Month {month} -> Graduated: {STUDENTS_GRADUATING_PER_MONTH}, Hired: {len(hired)}, Given up: {len(given_up)}, Remaining: {len(jobhunters)}")
  return all_hired, all_given_up, jobhunters

# Uncomment one of these three depending on what simulation you want.
hired, given_up, remaining = simulate_constrained_hires()
# hired, given_up, remaining = simulate_constrained_hires_with_improvement()
# hired, given_up, remaining = simulate_unconstrained_hires()
report(hired, given_up, remaining)