Lyft Data Scientist Interview Questions

Lyft·Data Scientist·Technical Phone Screen

Jul 2026

Lyft DS interview with a meaty case study spanning root cause analysis, pricing strategy, and experiment design all rolled into one question. It's the kind of thing that sounds manageable until you're 20 minutes in and realize you've barely scratched the surface.

Ride wait times at Lyft have increased 20% month-over-month. How would you structure a root cause analysis to systematically identify why, including defining KPIs and slicing by geography, time, and supply/demand dynamics?
How would you assess the costs and benefits of switching from static to dynamic pricing at Lyft, quantifying trade-offs across wait times, revenue, rider experience, and driver earnings?
Design an experiment to test whether dynamic pricing reduces wait times without hurting other key metrics, accounting for interference in a two-sided marketplace.

“I started with the obvious cuts: city, time of day, ride type.”

View Post

Lyft·Data Scientist·Technical Phone Screen

Jul 2026

Lyft data science interview with a probability/stats question built around a pricing experiment scenario. The setup was clever and required more careful modeling than it first appeared.

In a pricing experiment, each new rider is offered two rides on their first day, where each ride is priced reasonably with some independent probability P. What is the probability that the rider takes exactly one ride the following day? What is the probability they become a permanent rider who rides every day going forward?

“I stared at this for a second because the 'permanent rider' part felt open-ended.”

View Post

Lyft·Data Scientist·Technical Phone Screen

Jul 2026

Lyft data science interview with a meaty two-part case: diagnose a sudden metric change, then design an A/B test around the fix. The kind of question that sounds manageable until you're actually in it and realize how many moving parts there are.

A core business metric suddenly spikes or drops. Walk through how you'd systematically investigate the root cause.
Outline the key steps, assumptions, and success criteria for designing and analyzing an A/B test to validate a proposed fix for that metric issue.

“I started with time-based decomposition, which felt right, but I jumped to conclusions too fast about what segment was responsible before actually checking the data.”

View Post

Lyft·Data Scientist·Technical Phone Screen

Jun 2026

A deep system design question for a DS role at Lyft, all centered on building a dynamic pricing engine from scratch. The scope was massive and I'm not sure I did it justice, but it was a genuinely interesting problem to think through.

Design a dynamic pricing algorithm for a ride-hailing platform. Walk through the features and data sources you'd use, including real-time signals, historical patterns, and contextual factors like weather, traffic, and local events.
What model architecture would you use for this pricing system? Compare approaches like gradient-boosted trees versus reinforcement learning or contextual bandits, and explain the tradeoffs.
How would you formulate the objective function for this system and balance competing goals like rider conversion, driver earnings, company revenue, service quality, and price stability?
How would you incorporate fairness constraints, surge caps, price smoothing, and regulatory compliance into the pricing system?
How would you estimate price elasticity for this system? How do you handle causal identification given that prices aren't randomly assigned?
How would you validate this pricing system before and after launch, and how would you operate it reliably in production including monitoring, fallbacks, and serving infrastructure?

“I started with the obvious stuff (time of day, location density, weather) and then kind of spiraled trying to be comprehensive.”

View Post

Lyft·Data Scientist·Onsite - Multi Round

Jun 2026

Lyft data science onsite with a very heavy experimentation focus. The whole thing was essentially one long case about redesigning supplier prioritization in a two-sided marketplace, broken into five parts that kept building on each other. Tough but fair.

You're proposing a new supplier ranking policy to boost order completion in a two-sided marketplace with known interference between users and suppliers. How would you design the experiment, and which randomization unit would you choose given network effects and spillover risks?
What primary metric and guardrails would you define for this experiment, and at what level would you aggregate them?
Walk through the sample size calculation for this test. Baseline completion rate is 60%, minimum detectable effect is +2 percentage points absolute, two-sided test at alpha 0.05 and 80% power. Then adjust for a clustering design effect of 1.2.
How would you handle heterogeneous treatment effects across cities, time of day, and supplier capacity quartiles? And how do you run sequential monitoring without inflating Type I error?
How would you detect and correct for marketplace rebalancing effects, like treated geos improving at the expense of control geos?
Design a rollout ramp plan for this policy change and define a fallback decision rule if guardrail metrics breach for two consecutive days.

“This is where I spent the most time and also where I fumbled first.”

View Post

Lyft·Data Scientist·Technical Phone Screen

Jun 2026

Interviewed for a Data Scientist role at Lyft. One probability question that sounds straightforward until you realize you need to set up the conditional correctly.

A rider takes Lyft home. What's the probability they also took Lyft to work?

“I knew this was a Bayes setup but fumbled the direction of conditioning for a second.”

View Post

Lyft·Data Scientist·Onsite - Behavioral / Leadership

Jun 2026

Lyft data scientist interview that covered two pretty different areas: handling a vague, poorly guided case with a hard-to-understand interviewer, and a full comp negotiation scenario. Both felt more like situational judgment tests than anything technical.

Walk me through a time you worked on an open-ended project or interview where the other person had a strong accent and gave very little direction. How did you figure out what the actual problem was, set measurable success criteria, and avoid wasting time on things that didn't matter?
Describe a situation where your salary expectations didn't match what the role was offering. How did you prepare your range with data, handle the moment when they anchored low, and negotiate things like signing bonus or equity without damaging the relationship?

“This one was uncomfortable to answer because you have to be careful not to sound like you're complaining about the interviewer's accent.”

View Post

Lyft·Data Scientist·Onsite - Product Sense / Strategy

Jun 2026

Third round at Lyft for a Data Scientist role was a business case study, and I walked out pretty sure I'd bombed it. HR confirmed shortly after that they weren't moving me forward.

Prime Time pricing is stable across three consecutive weekends, then suddenly spikes in week 4. How do you investigate the root cause?

“I think I went too narrow too fast, jumped straight to demand-side stuff like a big local event and didn't structure it well before talking.”

View Post

Lyft·Data Scientist·Technical Phone Screen

Jun 2026

Lyft data science interview with a probability/Markov chain question built around their dynamic pricing feature. The setup was clever and felt more applied than the usual brain teaser, but it still required you to actually know your probability fundamentals under pressure.

Lyft is running a dynamic pricing trial. A rider takes two rides on day one, and each ride has probability P of being priced reasonably. Based on their day-one experience, what is the probability they take exactly one ride the next day? And what is the probability they become a permanent daily rider?

“The setup took me a minute to parse.”

View Post

Lyft·Data Scientist·Technical Phone Screen

Jun 2026

Lyft data scientist interview with a probability-heavy modeling problem built around a ride-hailing commuter scenario. Five parts, all chained together, and the whole thing lives or dies on whether you set up the per-trip independence correctly at the start.

A commuter takes two trips per day (morning and evening) with independent probabilities p and 2p of using the ride-hailing service. What is the probability the commuter takes at least one ride on a given day?
What is the probability that a commuter takes both the morning and evening rides on the same day, making them a 'committed' rider?
A rider uses the service on Day 1. What is the probability they take at least one ride again on Day 2, given that days are independent and follow the same model?
What is the probability that a new rider becomes a 'permanent' rider within two days, defined as being a committed rider (both trips) on at least one of Day 1 or Day 2?
Given that a rider took at least one ride on Day 2, what is the probability they took no evening (high-priced) ride on Day 1?
How would the answers change if morning and evening trip decisions within a day were positively correlated rather than independent?
Across N independent commuters, what is the distribution of the number who become committed riders on a given day? What are its mean and variance?
If you observed the empirical fraction of committed riders in real data, how would you estimate p and construct a confidence interval around it?

“Complement is the move here, 1 minus the probability of skipping both trips.”

View Post