← Back to Directory

Zoox interview notes, page 3 of 3

Mid-sized

Most of these Zoox write-ups come from Software Engineers (15), then Machine Learning Engineers (8) and Data Scientists (4). The most reported rounds are the technical phone screen (22), the system design round (4) and the behavioral / leadership round (1).

27 interview notes · updated Jul 2026

Zoox·Software Engineer·Technical Phone Screen

Jun 2026
SQL-heavy technical screen for a data engineer role at Zoox. Three questions, all on the same transaction/vendor schema, escalating from a basic ranking query to anomaly detection. Nothing behavioral, just SQL the whole time.
  • Given a transactions table and a vendors table, find the top 3 vendors by total net revenue (purchases minus refunds) over the past two years.
  • For each state and city combination, find the top 3 vendors by total net revenue over the past two years.
  • Propose several data quality and anomaly checks for this transaction dataset, then write SQL queries to detect each issue. Think about things like invalid refund references, mismatched refund amounts, duplicate transaction IDs, vendors not in the vendor table, and unusual spikes in volume or amount.

“Pretty standard warm-up.”

View Post

Zoox·Machine Learning Engineer·Technical Phone Screen

Jul 2026
Zoox ML engineer interview with a coding problem that looked like a simple scoring exercise but had a design follow-up that actually made me think. The OOP angle was the real test, not the math.
  • Given a list of roller coaster descriptions with type, max speed, bumps per second, and lift type, compute an overall score for each coaster using type-specific rules for scale factor and comfort score. Return the scores rounded to one decimal place.
  • Redesign your solution so that adding new coaster types or changing scoring rules doesn't require touching the core processing logic. How would you apply the open/closed principle here?

“The formula itself wasn't bad once I mapped out the three types separately.”

View Post

Zoox·Software Engineer·Technical Phone Screen

Jul 2026
Zoox SWE interview with a grid-based reachability problem. Clean problem once you see the trick, but I spent way too long overcomplicating it before the insight clicked.
  • An ice cream truck moves along a 2D integer grid and you're given its position at each time step. A second vehicle starts at the origin and can move one step in any cardinal direction or stay put each turn. Can the second vehicle ever occupy the same cell as the truck at the same time step?

“I went down a BFS path for a few minutes before realizing that was completely overkill.”

View Post

Zoox·Data Scientist·Technical Phone Screen

Jul 2026
Zoox data scientist interview with a classic systems-thinking question dressed up as a logistics puzzle. Nothing fancy but it required you to actually commit to tradeoffs rather than just listing options.
  • You need to get a 2 TB dataset to someone in San Francisco as quickly as possible. Walk through how you'd approach it and what method you'd pick, including any assumptions you're making about bandwidth, cost, security, or physical constraints.

“I went straight to network transfer and they pushed back almost immediately asking about realistic bandwidth.”

View Post

Zoox·Software Engineer·Technical Phone Screen

Jul 2026
SQL-heavy technical screen for a data engineer role at Zoox. Two questions, both involving the same transactions dataset, with the second one being a window function problem that I felt okay about but probably could've been cleaner.
  • Given a transactions table and a vendors table, write a SQL query to find the top 3 vendors by total purchase dollars in the last 2 years. Return vendor_id and total_dollars_charged, sorted highest to lowest.
  • Using the same dataset, for each (state_province, country) pair, find the top 3 vendors by total purchase dollars in the last 2 years. Return state_province, country, vendor_id, total_dollars, and vendor_rank. Use dense ranking so ties get the same rank.

“Pretty standard aggregation question.”

View Post

Zoox·Data Scientist·Technical Phone Screen

Jul 2026
Zoox data scientist interview that went deep into Bayesian inference, like really deep. The whole session was basically one extended probability problem with layers peeled back one at a time, which I was not fully prepared for.
  • You have a biased coin where one side lands with probability 2/3 and the other 1/3, but you don't know which side is which. Given equal priors, you flip it 4 times and get 3 heads and 1 tail. What is the posterior probability that heads is the heavier side?
  • Compute the Bayes factor comparing the two hypotheses given the observed data.
  • What is the posterior predictive probability that the next flip is heads, given the data you've already observed?
  • Generalize the setup to n flips with k heads. Derive a closed-form posterior, state the decision rule under 0-1 loss, and show that the Bayes factor simplifies to a power of 2.
  • Does the order in which the coin flips occurred matter under this model? Explain precisely.
  • How would the posterior and the predictive probability change if you used an asymmetric prior where P(heads-heavy) is not equal to 1/2?

“This is where it started and I actually felt okay here.”

View Post

Zoox·Software Engineer·Technical Phone Screen

Jul 2026
Data engineering interview at Zoox centered on a single meaty data quality question over a credit card transaction schema. Pretty technical, felt more like a working session than a traditional interview.
  • Given a credit card transactions table and a vendors table, propose at least five data quality checks or validations. Cover things like integrity constraints, referential integrity, refund logic, invalid values, missing data, and anomaly detection. For at least two of your checks, write SQL that returns the failing rows.

“This question is bigger than it looks.”

View Post