Snap Interview Questions — Page 7

Snap·Software Engineer·Technical Phone Screen

May 2026

Snapchat coding round, got a search autocomplete design question that's basically LC 642. More system-design-flavored than pure coding, which I wasn't fully expecting.

Design a search autocomplete system where, as a user types one character at a time, you return the top 3 historical sentences matching the current prefix, ranked by frequency and then alphabetically. Typing '#' finalizes the input and increments that sentence's frequency count. How do you keep the per-node top-3 list efficient?

“I went straight to a Trie and explained how each node could cache its top 3 sentences so you don't have to re-scan everything on each keystroke.”

View Post

Snap·Software Engineer·Onsite - System Design / Architecture

May 2026

System design round at Snapchat for a mid-level backend engineer role. The whole thing centered on designing a ride-hailing backend, which sounds broad but they kept pulling the conversation toward specific areas like matching logic and peak-load handling.

Design the backend for a ride-hailing platform like Uber. Walk through how ride requests are placed and matched with available drivers.
How would you handle a massive spike in ride requests during peak hours without the system falling over?
How would you store and serve rider and driver profile data at scale, and what consistency guarantees matter here?
Give a high-level overview of how you'd handle real-time driver location tracking on a map.

“I started with the API surface and request lifecycle, which felt right, but I fumbled when they pushed on the matching service.”

View Post

Snap·Software Engineer·Technical Phone Screen

May 2026

Snapchat software engineer interview that went deep on memory fundamentals. One question, but it had a lot of surface area and I felt like I was playing catch-up the whole time.

Can you walk through the differences between stack and heap memory? Cover how allocation and deallocation work, object lifetimes, performance, thread safety, what each region typically stores, and how Swift's value types and reference types map to each.

“This felt like five questions stitched into one.”

View Post

Snap·Machine Learning Engineer·Technical Phone Screen

May 2026

Snapchat ML Engineer interview that leaned hard into the practical side of model development. The imbalanced data question was the main technical focus and it went deeper than I expected.

How would you handle a severely imbalanced dataset in a classification task?

“I started with resampling (SMOTE, random oversampling, undersampling the majority) and class weighting in the loss function, which felt like the expected answer.”

View Post

Snap·Data Scientist·Technical Phone Screen

May 2026

Snapchat data scientist interview that went pretty deep into probability and stats. The coin flip setup sounds basic but they pushed all the way from expected value through geometric distributions to p-value reasoning, which I was not fully ready for.

For a single flip of a possibly unfair coin where heads has probability p, what is the expected value of the outcome? Follow-up: what is the variance?
If you flip a coin repeatedly until the first heads appears, what distribution does the number of flips follow, and what is the probability it takes exactly 3 flips?
What is the expected number of flips until the first heads in this geometric setup?
You ran the flip-until-heads experiment once and got T=3. How would you compute a p-value to test whether the coin is fair against the alternative that heads are rarer than expected?

“Pretty standard setup but the variance follow-up is where people slip up.”

View Post

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

May 2026

Snapchat ML Engineer system design round focused entirely on building a video recommendation system end-to-end. The depth they expected was real, covering everything from retrieval to ranking alignment in one session.

Design a video recommendation system end-to-end, covering candidate generation, ranking, and how the two stages stay aligned with each other.

“This one sprawled fast.”

View Post

Snap·Software Engineer·Technical Phone Screen

May 2026

Snapchat SWE interview with a pretty involved XML parsing question that covered tokenization, tree validation, mutation, and traversal all in one problem. A lot of moving parts for a single session and the complexity analysis expectations were no joke.

Implement a tokenizer for XML strings that classifies each token as an open tag, close tag, or raw text, and also accepts a pre-tokenized list as input.
Design an XMLParser class that takes a token list, validates the XML structure in O(n) time and O(h) space (where h is tree height), and raises exceptions for malformed input like mismatched tags, unclosed tags, or stray text.
Implement a to_string or __str__ method on the XMLParser that reconstructs the original XML from the internal representation.
Implement an add_element method that inserts a new child node under a node identified by a path, at a specified index, with optional text content.
Implement a remove_element method that deletes a node identified by a path from the tree.
Implement an iterative depth-first search traversal (no recursion allowed) that yields nodes in preorder using an explicit stack.

“Linear scan with a stack felt natural here.”

View Post

Snap·Machine Learning Engineer·Technical Phone Screen

May 2026

Snapchat ML Engineer interview that combined foundational ML concepts with a deep project walkthrough. You're expected to present a full ML project lifecycle and then defend every single choice you made under pretty aggressive follow-up. Not a casual conversation.

Walk me through a machine learning project you've worked on end-to-end, from how you framed the problem through to deployment and results.
Why did you make that specific technical decision in your project, and what alternatives did you consider?
What are the limitations of the approach you used, and what assumptions might not hold in production?

“This sounds open-ended but it's a trap in disguise.”

View Post

Snap·Software Engineer·Technical Phone Screen

May 2026

Snapchat SWE interview with a tree-based game theory problem that looks like a standard DP question until you realize the two-player simultaneous movement makes it way messier than expected.

You're given a rooted tree where each node holds some coins. Player 1 starts at the root and moves toward leaves one edge at a time, collecting coins. Player 2 picks any starting node and moves toward the root one edge at a time, also collecting. They move simultaneously. If they land on the same node at the same time, coins split (floor for Player 1). Game ends when Player 1 hits a leaf or Player 2 hits the root. Player 1 maximizes coins, Player 2 minimizes Player 1's total. Both play optimally. Return Player 1's coin count.

“I stared at this for a solid minute before I even understood the meeting condition.”

View Post

Snap·Software Engineer·Technical Phone Screen

May 2026

Snapchat SWE interview that leaned heavily on graph traversal. The core problem was classic islands but they pushed further than I expected, asking for a variant and a full comparison of three different approaches with complexity analysis.

Given an m-by-n grid where '1' is land and '0' is water, count the number of distinct islands using 4-directional connectivity.
Now modify the solution to return the sizes of all islands sorted in descending order instead of just the count.
Walk through DFS, BFS, and Union-Find approaches for this problem. Compare their time and space complexities and discuss when you'd prefer one over another.

“I went straight to DFS because it's the most natural fit and I can code it fast.”

View Post