← Meta Interview Insights

Meta·Data Scientist·Technical Phone Screen·Intermediate

Intermediate
Jun 2026

Summary

Meta data science interview with a probability/statistics question that leans on the hypergeometric distribution. One question, fairly niche, not your typical SQL or A/B testing warmup.

Questions Asked (1)

Q1

There are 1,000 employees and 10 are randomly selected each round for a perk. On average, which round would you expect to be selected for the first time?

Algorithms & Data StructuresProduct Analytics & Metrics
Author's notes

I knew it smelled like a hypergeometric setup but blanked on the exact expected value formula mid-explanation.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Model the selection process as a geometric distribution where each round is an independent Bernoulli trial with success probability p = 10/1000 = 0.01. The expected number of rounds until first success is 1/p = 100, so on average, you would be selected in the 100th round. Clearly state assumptions (random selection with replacement, independence) and explain the memoryless property.

Pro tip: Mention that the expected round is 100, but the distribution is highly skewed: the median is about 69 rounds, and there's a 36.6% chance of waiting more than 100 rounds. This shows deeper understanding and avoids the common misconception that the average is the most likely outcome.

1. Define the random variable

Let X be the round number of your first selection. Each round, the probability of being selected is p = 10/1000 = 0.01, assuming independent draws with replacement.

2. Identify the distribution

X follows a geometric distribution with success probability p = 0.01, since we are waiting for the first success in a sequence of independent Bernoulli trials.

3. Compute the expected value

The expected value of a geometric distribution is E[X] = 1/p = 1/0.01 = 100. So on average, you expect to be selected in the 100th round.

4. Discuss variability and implications

Mention that the distribution is right-skewed: the median is about 69 rounds, and the probability of waiting more than 100 rounds is (0.99)^100 ≈ 0.366. This highlights that the average may not represent a typical wait time.

Key Points to Mention

  • Geometric distribution and its expected value formula E[X] = 1/p
  • Independence of rounds and constant probability p = 0.01
  • Memoryless property: past rounds do not affect future probabilities
  • Median vs. mean: median is approximately 69 rounds, showing skewness
  • Probability of waiting more than 100 rounds is about 36.6%
  • Assumption of with replacement (or large population) ensuring constant p

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.