Airbnb Data Scientist Interview Questions

Airbnb·Data Scientist·Technical Phone Screen

Jul 2026

Airbnb data scientist interview that was basically a full take-home disguised as a technical screen. One massive pandas question covering sessionization, funnel analysis, segment metrics, and statistical lift calculations. A lot to fit into one sitting.

Given a raw events table and a guests table, write idiomatic vectorized Pandas to: sessionize events per guest using a 30-minute inactivity timeout, filter bot traffic, then compute segment-level metrics (unique guests, sessions, session-level and guest-level conversion rates, average sessions-to-first-order, and median time from first page_view to first order_completed) broken out by traffic_source and device. Restrict the analysis to a 7-day window ending 2025-09-01.
Using the same event data, build a de-duplicated guest funnel for page_view, add_to_cart, checkout_start, and order_completed over the 7-day window. Each guest should be counted once at their highest attained step. Report step-through rates and absolute drop-offs.
After filtering segments with fewer than 100 sessions, identify the top 5 segments by lift versus the sitewide guest-level conversion rate. Report both the lift and the 95% Wilson confidence interval for each segment's conversion rate.
How do you handle guests who complete an order_completed event in the window but have no earlier events in the same window? Implement a specific rule: pull their first event from the prior 24 hours if available, otherwise treat them as a direct single-step conversion.
How would you adapt this Pandas pipeline to run at scale in PySpark? What are the key translation points?

“The sessionization part is where I spent most of my mental energy.”

View Post

Airbnb·Data Scientist·Technical Phone Screen

Jul 2026

Airbnb data science interview with a pretty gnarly causal inference scenario. No fluff, just one meaty question about estimating impact when you've already blown past the experiment window.

A feature was launched to 100% of users with no control group. How would you estimate the causal impact of that launch, what data would you need, what assumptions are you making, and how do you communicate the uncertainty?

“This one stung a little because I started with pre-post analysis and the interviewer just waited.”

View Post

Airbnb·Data Scientist·Technical Phone Screen

Jun 2026

Airbnb data scientist interview that dropped me into a product-planning scenario where I had to define the full experiment framework before a single line of code gets written. Pretty rigorous, covers a lot of ground in one question.

You're in a product planning session before development starts. Walk through the metrics framework, guardrails, experiment design, unit of randomization, sample size calculation, and how you'd monitor the experiment while it's running.

“This is a lot to hold in your head at once and I think I fumbled the ordering.”

View Post

Airbnb·Data Scientist·Take-home Assignment

Jun 2026

Airbnb data scientist interview with a meaty take-home style prompt: raw A/B test logs, full pipeline from cleaning to stakeholder communication. Felt like a real work sample more than a quiz, which I actually appreciated.

You're given raw log-level data from an A/B test. Clean it in Python, compute the key metrics (conversion rate, lift, p-value), build at least one visualization, and explain what the result means for the business.

“This is the kind of question that sounds manageable until you're staring at messy logs and realize half your events are duplicated or missing timestamps.”

View Post

Airbnb·Data Scientist·Technical Phone Screen

Jun 2026

Airbnb data science interview with a meaty experimentation question that felt more like a case study than a behavioral. The kind of question where you realize halfway through your answer that you're being tested on three different things at once.

A major feature launched globally without a holdout group and stakeholders want a fast read on whether to keep it or roll it back. How do you approach this, what alternatives do you propose, and how do you get skeptical PMs, engineers, legal, and marketing to align on a path forward?
How would you communicate uncertainty and statistical risk to non-technical stakeholders when there's pressure to make a fast ship or rollback call?
If the team culture were highly academic and passive, what would you do differently in terms of driving alignment and moving toward a decision?

“This one sprawled in a way I wasn't prepared for.”

View Post

Airbnb·Data Scientist·Technical Phone Screen

May 2026

Airbnb data science interview with a single very deep experimentation question. The whole thing was basically a case study on designing an A/B test for a new Wi-Fi badge feature, and it went several layers deeper than I expected.

You work on a two-sided travel search marketplace. Product wants to add a 'High Wi-Fi' badge and filter to help remote workers find listings. Design an experiment to decide whether to launch it, covering randomization strategy, metrics, power analysis, analysis plan, and a dollar-value decision rule.

“This was a monster.”

View Post

Airbnb·Data Scientist·Technical Phone Screen

May 2026

Airbnb data scientist interview that went deep into metric design and experimentation rigor. The question was essentially one long technical prompt covering everything from logging integrity to power calculations, and it felt less like a conversation and more like defending a thesis.

For a newly launched feature, define a primary success metric tied to long-term value. Walk through the exact formula including numerator/denominator definitions, deduplication logic, windowing choices, and how you handle outliers and bot traffic.
What are the trade-offs between ratio-of-sums and sum-of-ratios when computing your primary metric, and how does your choice affect sensitivity to late or backfilled events?
How would you detect and correct a silent logging change that occurs mid-experiment, and what does timezone misalignment do to your metric computation?
Given historical variance from past experiments, compute the minimum detectable effect and required sample size for a 14-day test. Then explain how you'd approach sequential monitoring and alpha spending across that window.
Define at least three guardrail metrics for the experiment, set canary thresholds for each, and explain what you'd do if the primary metric improves but one guardrail shows a slight regression.

“This is where I spent most of my time and also where I stumbled first.”

View Post

Airbnb·Data Scientist·Take-home Assignment

May 2026

Airbnb DS interview that was basically one big SQL/Python case study disguised as a take-home. The problem looked manageable at first glance until I actually read the attribution rules and edge cases, then it became a whole thing.

Given a visits table and a bookings table, deduplicate visits within a 60-second window per user (keep only the latest), then attribute each confirmed booking to the most recent visit by the same user within the prior 24 hours. Compute control vs treatment metrics by day and variant for a 7-day window, including visits, unique users, attributed bookings, conversion rate, total revenue, revenue per visit, cancellation counts, and unattributed confirmed booking counts. Provide SQL or Python and explain your assumptions.
Produce a daily line plot of conversion rate for control and treatment with 95% Wilson confidence intervals.
How do you handle edge cases like a user appearing in both control and treatment within the same 7-day window, multiple bookings following a single visit, or a booking that falls just outside the 24-hour attribution window?

“The deduplication step is where I stumbled first.”

View Post

Airbnb·Data Scientist·Onsite - Cross-functional / Panel

May 2026

Cross-functional panel at Airbnb for a data scientist role, mix of product and engineering folks in the room which made it feel less predictable than a standard behavioral loop.

Tell me about a time you influenced a decision without having direct authority over the people involved. How did you get everyone aligned, and what actually came of it?

“I went in thinking this was a softball and kind of rambled through a story without anchoring it to any real numbers.”

View Post

Airbnb·Data Scientist·Technical Phone Screen

Apr 2026

Airbnb data scientist interview that was basically a three-part gauntlet: a gnarly SQL panel-building problem, a rolling-window metric question, and then a Python DiD estimation task that went pretty deep into econometrics. The technical bar was higher than I expected, especially the fixed-effects stuff at the end.

Given tables for users, exposures, and orders, build a user-day panel covering a specific date range. Each row should include whether the user was treated, whether their country had launched yet, their revenue for that day, whether they were active, and how many days since signup. Handle missing user-days and avoid duplicates.
Using the panel you built, compute a 7-day rolling revenue per active user for two specific countries over a defined date window. Explain how you handle users with no activity and how you avoid lookahead bias.
Using the panel, estimate a two-way fixed-effects difference-in-differences model in Python with user and day fixed effects and clustered standard errors. Then: interpret the interaction term economically, test for pre-trends using leads, discuss late adopter and immortal time bias, re-estimate using a Poisson count model with an exposure offset, and give a 95% confidence interval plus a practical recommendation.

“This took way longer than it should have.”

View Post