Dataford
Popular roles
Software EngineerData AnalystData ScientistData EngineerBusiness AnalystAI EngineerMachine Learning EngineerProduct Manager
Browse
Browse All RolesEvery role hub, from analyst to MLBrowse All CompaniesCompany-specific interview loopsAll Interview GuidesThe full guide library
Top questions by role
Software EngineerData AnalystData ScientistData EngineerBusiness AnalystAI EngineerMachine Learning EngineerProduct Manager
Top questions by skill
SQLPythonStatisticsMachine LearningA/B TestingSystem DesignGenerative AIProduct SenseMetricsBehavioral
Browse all questions →Try a mock interview
ExperiencesCoaching
New
Practice
Mock InterviewsTimed interview simulations with feedbackSuccess PathYour 6-week structured planModulesCurated lessons by topicWebinarsTalks from ex-Big Tech data leadsPlaygroundA free-form scratch editor
Learn
BlogInterview strategy and career adviceTech Job Market ReportHiring trends across data and AI rolesFor UniversitiesDataford for career centersAbout DatafordWho we are and how we build
Pricing
Get started

Bayesian Coin Inference for Reels

EasyStatistics & Probability00:00
Practice interviewer
In session
5 left
00:00

Whenever you're ready, walk me through how you'd approach this.

You need to log in / sign up to chat or submit.

Problem

Business Context

A Meta DS is sanity-checking a toy Bayesian reasoning question used in onboarding for experimentation analysts working on Instagram Reels. Think of the coin as a hidden traffic source: one source behaves like a fair process, and one source always produces a positive signal. After observing one positive outcome, you need to infer which source was more likely selected.

Problem Statement

You randomly pick 1 of 2 coins with equal probability:

  • Coin F is a fair coin
  • Coin H is a double-headed coin

You flip the selected coin once and observe Heads. Determine the probability that you had picked the fair coin.

This is the same conditional-probability logic Meta DSs use when reasoning about posterior probabilities after seeing an observed event, such as an unexpectedly high early IG Save rate in a small Reels slice before checking for SRM, novelty effect, or whether CUPED-adjusted priors change interpretation.

Given Data

QuantityValue
Number of coins2
Prior probability of choosing fair coin0.50
Prior probability of choosing double-headed coin0.50
Probability of Heads given fair coin0.50
Probability of Heads given double-headed coin1.00
Observed outcomeHeads

Requirements

  1. Define the prior probabilities and likelihoods.
  2. Use Bayes' theorem to compute the posterior probability of having picked the fair coin given that Heads was observed.
  3. Show the total probability of observing Heads.
  4. Briefly interpret why the answer is not 0.50 after seeing Heads.

Assumptions

  • The coin is chosen uniformly at random.
  • The observed flip is accurate.
  • Only one flip is observed.
  • No AARRR funnel segmentation, CUPED adjustment, or SRM issue affects the setup.