Snowflake Interview Questions — Page 13

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

Jun 2026

Snowflake data scientist loop, heavy on the case-study presentation format. You bring a real shipped project and they pick it apart for about an hour across metrics, mistakes, stakeholder drama, and cross-functional process.

Walk us through an end-to-end project you led that shipped to users: the problem, stakeholder goals, data sources, modeling approach, key decisions, results, and trade-offs. You have 10 minutes and up to 5 slides.
What was the single metric you optimized for, what guardrails did you put in place, and why? Tell us about a time your chosen metric conflicted with another stakeholder's metric and how you resolved it.
Describe one concrete mistake from the project: a wrong metric trade-off, a flawed assumption, anything that actually hurt the outcome. What did you change afterward?
Leadership pushes back on your proposal because the metrics are moving in opposite directions, say retention is up but revenue is down. How do you redesign the follow-up experiment or rollout to address that without blowing up the timeline?
Give a specific example of working with data engineering, product, or design under time pressure. How did you negotiate scope or data model changes when you didn't have the luxury of time?

“This is the anchor for the whole session so whatever you pick sets the ceiling for every follow-up.”

View Post

Snowflake·Software Engineer·Technical Phone Screen

Jun 2026

Snowflake software engineer interview with a grid pathfinding problem and a follow-up on DFS vs BFS tradeoffs. Pretty standard algorithmic round but the follow-up is where things got interesting.

Given an m x n grid of open and blocked cells, find the minimum number of moves to get from a start cell to a target cell moving in four directions. Return -1 if unreachable.
How would you solve the same grid pathfinding problem using recursive DFS with backtracking and memoization, and what are the tradeoffs versus BFS?

“Classic BFS setup, went fine.”

View Post

Snowflake·Software Engineer·Onsite - System Design / Architecture

Jun 2026

Snowflake system design round for a software engineer role. The whole thing was one long question about building a dependency-aware service scheduler, and they kept pushing deeper into every corner of it.

Design a scheduler that starts services on a host with a fixed number of CPU cores, where services have dependencies forming a directed acyclic graph. The goal is to minimize total startup time while respecting those dependencies. Walk through how you detect which services are ready to run, how you maximize parallelism without exceeding core limits, how you handle prioritization, timeouts, retries, failures, and backoff. Also cover resource constraints beyond CPU (memory, ports), observability, the data structures you'd use, correctness guarantees, and how you'd extend this across multiple machines.

“This is the kind of question that looks manageable for the first five minutes and then just keeps expanding.”

View Post

Snowflake·Software Engineer·Technical Phone Screen

Jul 2026

Snowflake software engineer coding round with a graph traversal problem. Pretty clean problem once you figure out what they're actually asking, but the setup takes a second to parse.

You're given a list where each index represents the privileges belonging to a role, and a list of grants where one role inherits from another (directed edges). Compute the full set of privileges each role ends up with after inheriting from all its ancestors.

“First thing I did was ask if cycles were possible.”

View Post

Snowflake·Software Engineer·Technical Phone Screen

Jul 2026

Snowflake coding interview with a string manipulation problem that looks deceptively simple but has a few layers to it. The palindrome-via-swaps angle threw me a bit.

Given a string containing only two distinct characters, determine if it can be rearranged into a palindrome using only adjacent swaps. If yes, return the minimum number of adjacent swaps needed; otherwise return -1.

“I got the feasibility check pretty fast (odd-length strings can have one character with an odd count, even-length need both even) but the minimum swap part is where I started fumbling.”

View Post

Snowflake·Software Engineer·Technical Phone Screen

Jul 2026

Snowflake software engineer interview with a pretty focused algorithmic problem around binary trees. The question had enough layers to it that I kept second-guessing my approach the whole time.

Given two complete binary trees with identical structure, update each node in the second tree so its value equals the sum of all values in the subtree rooted at the corresponding position in the first tree. The solution must run in linear time and use at most O(h) extra space where h is the tree height.

“My first instinct was a straightforward recursive postorder traversal, compute subtree sums in tree one and assign them to tree two simultaneously.”

View Post

Snowflake·Software Engineer·Onsite - System Design / Architecture

Jul 2026

System design round at Snowflake for a software engineer role, centered entirely on one gnarly distributed systems problem. The scope was massive and I definitely didn't cover everything they wanted, but it was a genuinely interesting problem to think through.

Design a global, multi-tenant quota service where multiple upstream products share a single per-user quota. The system must support consuming, releasing, and querying quota, while never allowing overdraft, and must scale to millions of daily active users.

“This one ate me alive a little.”

View Post

Snowflake·Software Engineer·Onsite - System Design / Architecture

Jul 2026

System design round at Snowflake for a software engineering role. The whole session was basically one big crawler question with a lot of follow-ups packed in, and they clearly wanted to see if you'd thought through the messy real-world stuff, not just the happy path.

Design and implement a web crawler that starts from a given URL, uses an interface to fetch outgoing links, and returns all pages under the same hostname without revisiting any URL. You also need to support a configurable concurrency limit.
How do you ensure thread-safe deduplication when multiple workers are crawling concurrently?
How would you handle URL normalization to avoid treating the same page as different URLs?
How would you implement politeness in the crawler, including rate limiting and respecting robots.txt rules?
How would you handle errors during crawling, such as failed fetches or unreachable pages?
How would you test this crawler for both correctness and performance?

“This is the kind of question that feels manageable until you actually start talking through it.”

View Post

Snowflake·Software Engineer·Technical Phone Screen

Jul 2026

Quick Snowflake SWE coding round, just one problem but with a twist thrown in at the end that I wasn't fully prepared for.

Given a list of intervals that are NOT sorted in increasing order, merge or process them accordingly (a variant of the classic interval problem with unsorted input).

“The interviewer just dropped this on me right after the previous problem, no warmup.”

View Post

Snowflake·Software Engineer·Technical Phone Screen

Jul 2026

Snowflake software engineer interview with a graph traversal problem. Pretty standard BFS setup but the edge cases tripped me up more than I expected.

Given a start and target wiki page URI, implement a function that returns the minimum number of clicks (link hops) to get from one to the other. You have access to an API that returns all outgoing links from a given page. Return 0 if they're the same page, -1 if unreachable.

“My first instinct was DFS and I'm glad I caught myself before coding it up because that would've been wrong.”

View Post