Snap Interview Questions — Page 4

Snap·Software Engineer·Technical Phone Screen

Apr 2026

Snapchat SWE interview with a product co-purchase frequency problem. Pretty algorithmic but had a product data feel to it, like something you'd actually build for a recommendation system.

Given a list of orders where each order is a list of product IDs bought together, and a query product ID, return all products co-purchased with the query product at least once, sorted by co-purchase frequency descending and by product ID to break ties.

“The co-occurrence map part was fine, just iterate through orders and for each order containing the query product, increment counts for the other items.”

View Post

Snap·Software Engineer·Technical Phone Screen

Apr 2026

Snapchat coding round focused on binary tree traversals, which sounds chill until you realize the real test is getting the output format exactly right. The algorithm part was fine but the formatting details ate up more time than expected.

Given a binary tree, implement a traversal (level-order, vertical, or boundary order as specified) and print the result in a precise format such as one line per level, comma-separated values, or a serialized string.

“The traversal logic itself wasn't the hard part.”

View Post

Snap·Software Engineer·Onsite - Behavioral / Leadership

Apr 2026

Behavioral round at Snapchat for a software engineering role. One meaty question that took up most of the time, covering a lot of ground about how you handle pressure and trade-offs. Left feeling like I either nailed it or completely rambled my way through it.

Tell me about a time you had to deliver a project under serious time pressure. Walk me through how you scoped the work, what you cut, how you kept stakeholders informed about the trade-offs, how you kept quality from falling apart, and what the outcome was. What would you do differently?

“This question is doing a lot of work at once and I didn't clock that fast enough.”

View Post

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

Apr 2026

Snapchat MLE system design round, full hour on video recommendation. No coding, just architecture and trade-offs the whole time. Felt pretty solid in some areas and completely exposed in others.

Design a video recommendation system for a short-video or YouTube-style platform. Walk through the full architecture from candidate retrieval to final serving.
What are the non-functional requirements you'd prioritize for a recommendation system at this scale, and how do they affect your design choices?
How would you design the data pipeline for logging user interactions and turning that into training data for the recommendation model?
How do you handle storage and indexing for a large item corpus, and how does that interact with your retrieval latency requirements?
How would you set up A/B testing for a recommendation model change, and what metrics would you track?

“This was the whole interview basically.”

View Post

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

Apr 2026

System design round at Snapchat for an MLE role, centered entirely on one very long product tagging question. The scope was massive and I kept second-guessing whether to go broad or deep. Came away feeling like I handled the ML modeling parts okay but fumbled the downstream and adversarial sections.

Design a full end-to-end pipeline that automatically assigns category, attribute, and quality tags to products on a short-video commerce platform, starting from seller-uploaded titles, descriptions, images, and video.
How would you design the taxonomy and label space for a product tagging system, and how does that choice affect your modeling approach?
Walk through how you'd handle weak supervision and human-in-the-loop labeling to bootstrap and maintain training data for this kind of tagging system.
What are the tradeoffs between running this tagging pipeline as an offline batch job versus a real-time online service triggered at listing creation or update?
How would you monitor tag quality in production, specifically tracking precision and recall per tag and detecting model or data drift over time?
How do the product tags your system generates get consumed by downstream systems like search, recommendations, and ads, and does that affect how you design the tags themselves?
How would you handle multilingual product catalogs, and what would you do about adversarial sellers who try to game the tagging system by stuffing misleading keywords or manipulating images?

“This question ate the entire session.”

View Post

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

Apr 2026

One-hour system design round at Snapchat for an MLE role, focused almost entirely on building a TikTok-style short video recommendation system end to end. Pretty intense scope for a single hour, covering everything from retrieval to online inference to training pipelines.

Design a short-video recommendation system (TikTok-style) with emphasis on the distributed system architecture for both model training and online serving.
How would you design the feature store to handle user features, video features, and real-time interaction signals at scale?
Walk through how you'd set up the distributed training pipeline, including your choice of parameter server versus all-reduce and how you'd handle incremental or online learning.
How would you approach AB testing and experimentation infrastructure for a recommendation system at this scale?
What are the scaling challenges when moving from millions to billions of videos and hundreds of millions of active users, and how does your design address them?

“This was the whole interview.”

View Post

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

Apr 2026

Snapchat MLE interview focused entirely on ads ranking for the search surface. No infrastructure deep-dive, but the scope was wide: query understanding, retrieval, ranking, auction mechanics, blending with organic, freshness, and both offline and online eval. Pretty dense for a single session.

How would you design an ads ranking system for the search surface, covering the full pipeline from query understanding through auction and blending with organic results?
What features and training objectives would you use for the ad ranking model?
How would you handle freshness in the ads ranking system?
How would you evaluate the ads ranking system both offline and online?
How would you design the pricing and auction mechanism for search ads?
How would you blend ads with organic search results in a way that feels natural to users?

“This is the kind of question where you can either anchor on one layer and go deep, or try to sketch the whole thing and stay shallow everywhere.”

View Post

Snap·Software Engineer·Technical Phone Screen

Apr 2026

Snapchat coding interview with a grid-based reachability problem that started simple and then got way more interesting once they added the dynamic blocker twist. The follow-up about incremental data structures is where things got real.

Given a grid with a start, an end, and some blocked cells, determine if the end is reachable from the start. Then: a new blocker gets added to the grid. How do you handle reachability efficiently going forward? Walk through the trade-offs between recomputing from scratch versus maintaining some incremental structure.

“The BFS part I got through fine, pretty standard.”

View Post

Snap·Software Engineer·Technical Phone Screen

Apr 2026

Snapchat SWE interview that was basically a systems design session dressed up as a coding round. Started with a clean pub/sub implementation and then kept layering on follow-ups until we were deep in concurrency territory. Pretty solid interview if you like that kind of incremental design problem.

Design and implement an in-memory Pub/Sub system with createTopic, subscribe, and publish operations.
Modify the publish method so subscribers receive messages via push delivery instead of polling, using callbacks or per-subscriber queues.
How would you handle concurrent publishers and subscribers in this system? Walk through a mutex-based approach and then explain how a read/write lock improves throughput.

“The base implementation wasn't too bad.”

View Post

Snap·Software Engineer·Technical Phone Screen

Apr 2026

Snapchat SWE interview with a grid-based pathfinding problem. Pretty classic BFS territory but the multi-source, multi-person angle adds a layer that tripped me up a bit on the complexity analysis.

You're given a 2D grid with walls, people, and exits. Can every person reach an exit? If yes, what's the minimum time for all people to escape, where each person moves optimally and time is the max shortest-path distance across all people?

“My first instinct was to BFS from each person separately to find their nearest exit, then take the max.”

View Post