Glean Machine Learning Engineer interview questions

7 Machine Learning Engineers have shared notes from their Glean interviews, with the questions they were asked in each round. That is 26% of the 27 Glean interview notes on Screna AI.

7 Machine Learning Engineer interview notes

Glean·Machine Learning Engineer·Technical Phone Screen

Jul 2026
Interviewed for an MLE role at Glean and got a graph/search problem dressed up as a train scheduling puzzle. Took me a minute to see it wasn't just a simple BFS on a static graph because time actually matters here. Decent problem, probably mid-to-hard difficulty.
  • Given a train schedule where each route is a list of stations indexed by absolute time, write a function to determine if a passenger starting at a given station at time 0 can reach a destination station.
  • How would you model the state space for this train reachability problem, specifically accounting for waiting at stations?
  • What is the time complexity of your solution, and what edge cases would you need to handle?
  • How would you modify the solution to return the actual path taken, not just a boolean?
  • What changes if trains have arbitrary timestamps instead of route-index-based time?

“The core trap I nearly fell into was treating this like an undirected graph problem.”

View Post

Glean·Machine Learning Engineer·Onsite - System Design / Architecture

Jun 2026
Did an ML design round at Glean for an MLE role. One question, 60 minutes, focused on computing distances between employees. Interesting problem space but the session left me with a lot of unanswered questions about what they actually wanted.
  • Design a system that computes a distance or similarity score between any two employees in an organization.

“This is the kind of question that sounds deceptively scoped until you realize you need to define what 'distance' even means before writing a single line of pseudocode.”

View Post

Glean·Machine Learning Engineer·Take-home Assignment

May 2026
Glean gave me a 75-minute AI-assisted coding assignment to build a Wikipedia crawler with some specific prioritization logic. The tool choice felt off and I spent most of the session trying to decipher code I didn't fully trust.
  • Build a Wikipedia crawler that prioritizes pages whose title starts with a letter not yet visited. Once all 26 letters are covered, fall back to random crawling. Must respect a rate limit. You can use AI coding tools.

“The logic itself isn't that bad once you map it out.”

View Post

Glean·Machine Learning Engineer·Technical Phone Screen

May 2026
Coding round for an MLE role at Glean. The problem was building a rate-limited Wikipedia crawler with some prioritization logic baked in. AI coding tools were allowed, which was a nice change, though you still had to own and explain everything you wrote.
  • Design and implement a rate-limited Wikipedia crawler that starts from a given page, extracts outgoing links, and prioritizes crawling pages whose starting letter hasn't been visited yet. Once all 26 letters are covered, fall back to random selection from the frontier. Handle deduplication, configurable rate limits, and common failure cases.

“The alphabetical prioritization was the part I kept second-guessing myself on.”

View Post

Glean·Machine Learning Engineer·Onsite - System Design / Architecture

May 2026
ML system design round at Glean for an MLE role. The whole interview was a single deep-dive on building an employee similarity/distance system, which sounds contained but sprawls fast once you get into privacy, scale, and the five different use cases that all want different things from the same score.
  • Design an employee-to-employee distance system for a large company. It should power people search, internal networking recommendations, collaboration discovery, org insights, and onboarding suggestions. Walk through what 'distance' means, what data you'd use, how you'd model it, how you'd serve it at scale, and how you'd handle privacy and fairness.
  • A new hire has almost no collaboration history. How does your system avoid returning a useless result where everyone looks equally far away on day one?
  • Two different viewers querying the same pair of employees should potentially see different distance scores based on what each viewer is authorized to know. How do you make that work without recomputing everything per viewer?
  • Same-team employees almost always dominate the top of recommendation lists. How would you detect this over-ranking and actually surface useful cross-functional connections?
  • Low-cardinality signals like 'two co-attended meetings' can reveal who met with whom even if you never expose raw calendar data. How do you prevent that kind of inference attack?

“I started by trying to collapse everything into one score and the interviewer pushed back almost immediately, which was fair.”

View Post

Glean·Machine Learning Engineer·Technical Phone Screen

Apr 2026
Interviewed for an MLE role at Glean and got a graph/search problem that looked like a scheduling puzzle. Two parts, progressively harder, and the second part tripped me up more than I expected.
  • You're given a set of train itineraries, each as a sequence of (station, time) stops. Given a start station, start time, and destination, find the earliest possible arrival time at the destination.
  • Extend your solution to reconstruct and print the full route: every station visited, the time, which train was taken, any waits, and any transfers between trains.

“My first instinct was Dijkstra but then I realized the waiting mechanic made it weird.”

View Post

Glean·Machine Learning Engineer·Technical Phone Screen

Apr 2026
Glean ML engineer interview with a graph traversal problem disguised as a transit scheduling puzzle. Clean problem statement but the follow-ups are where it gets interesting.
  • Given a train schedule where each train is a list of stations (index = arrival time), implement a function that determines whether a passenger can travel from a start station to an end station. Passengers can transfer between trains at shared stations even if they arrive at different times, since they can wait.

“The waiting mechanic is the part that trips people up.”

View Post