← Back to Directory

Netflix interview notes, page 16 of 20

FAANG / Big Tech

Most of these Netflix write-ups come from Software Engineers (136), then Data Scientists (23) and Product Managers (16). The most reported rounds are the technical phone screen (94), the system design round (48), the product sense round (14) and the behavioral / leadership round (11).

191 interview notes · updated Jul 2026

Netflix·Software Engineer·Technical Phone Screen

Jun 2026
Netflix SWE interview with a sliding window problem dressed up in streaming flavor. Pretty standard underneath the theme.
  • Given a list of show names being streamed in sequence, find the length of the longest contiguous subsequence where no show name repeats.

“It's basically the classic sliding window substring problem but with show titles as the elements.”

View Post

Netflix·Software Engineer·Onsite - System Design / Architecture

Jun 2026
Netflix system design round for a software engineer role. The whole session basically revolved around one big cache design problem, but it kept branching out into memory, concurrency, and distributed systems territory before I even felt settled on the core answer.
  • Design a cache class with get and put operations. What eviction policy would you use, and how would you implement it efficiently?
  • How would you handle memory management in your cache? Think about entry sizing, soft caps, and when eviction should actually trigger.
  • How would you make your cache thread-safe? Walk through the tradeoffs between coarse-grained locking, fine-grained locking, and lock-free approaches.
  • How would you scale this cache across multiple machines? Consider sharding, hot key handling, and cache stampede prevention.

“I went straight to LRU because it felt like the expected answer, and it is, but I fumbled the explanation of why a doubly-linked list plus a hashmap gives you O(1) on both ends.”

View Post

Netflix·Software Engineer·Onsite - System Design / Architecture

Jun 2026
Netflix system design round focused entirely on ads infrastructure. One big open-ended question that kept branching into deeper territory the longer we talked.
  • Design the data model for an ads demand-intake system, walking through the full hierarchy from account down to impression events, including entities, attributes, and relationships.
  • How does your data model support targeting logic, specifically across geo, device type, audience segments, and frequency caps?
  • Walk through how your model handles bidding, including bid types, pacing, budget caps, and the difference between first-price and second-price auctions.

“I started at the impression level and worked upward, which in hindsight was the wrong call.”

View Post

Netflix·Software Engineer·Technical Phone Screen

Jun 2026
Netflix data engineer screen, one meaty coding problem about streaming event processing. The problem itself was interesting but the follow-up tripped me up more than I expected.
  • Given a stream of user-movie viewing events, each containing a user ID, movie ID, and percent watched, deduplicate by keeping the maximum percent watched per user-movie pair, then classify each pair into completion buckets (e.g. 30%, 50%, 80%) and return the count of pairs in each bucket.
  • Sort the buckets by count in descending order and return just the bucket names, not the counts.

“The dedup part clicked pretty fast.”

View Post

Netflix·Software Engineer·Technical Phone Screen

Jun 2026
Netflix coding round for a Software Engineer role, second in a three-part progressive problem set. The question was a twist on a classic sliding window problem, scaled up to work on arrays of strings instead of individual characters.
  • Given an array of strings (like a list of names), find the longest contiguous subarray where every string is unique. Return either the length or the subarray itself.

“I recognized the sliding window pattern pretty fast since it's basically the same idea as the single-string version.”

View Post

Netflix·Data Scientist·Technical Phone Screen

Jun 2026
Netflix data scientist interview with a meaty system design question around fraud detection. Single question but it covered a lot of ground, from feature engineering all the way to post-deployment monitoring.
  • Walk through the full pipeline for training and deploying an XGBoost model to detect fraudulent subscription payments in real time. What evaluation metrics matter most, how do you deal with heavy class imbalance, and how would you catch model drift after the thing is live?

“This question is basically four questions stitched together and I felt it the whole time.”

View Post

Netflix·Software Engineer·Onsite - Coding / Algorithms

Jun 2026
Netflix SWE coding round, two problems back to back. The cache one was manageable but the tree problem had a subtle O(n) constraint that tripped me up a bit.
  • Design and implement an in-memory key-value cache where each entry expires after a given time-to-live. You need to support put, get, and a count of currently live keys, all taking a current timestamp as a parameter.
  • Given a binary tree, output each node's value, its depth level, and whether it's balanced (left and right subtree heights differ by at most 1). Do this in preorder traversal and make sure the whole thing runs in O(n) without recomputing heights multiple times.

“Felt pretty solid on this one.”

View Post

Netflix·Software Engineer·Technical Phone Screen

Jun 2026
Netflix coding screen, one problem the whole session. The question was about deduplication logic for a streaming app home page shelf layout. Took me a while to fully internalize the two-phase dedupe rules before I could even start writing anything.
  • Given a list of shelves (each shelf is a list of title IDs) and a viewport width X, filter each shelf's titles according to a two-phase deduplication rule: for the first X output titles on any shelf, skip titles already seen in the visible region of previous shelves (global dedupe); beyond position X on a shelf, only skip titles that already appeared within that same shelf (local dedupe). Return the filtered shelves preserving original order.

“The problem sounds manageable until you sit with the phase boundary.”

View Post

Netflix·Software Engineer·Online Assessment (OA)

Jun 2026
Netflix SWE OA, pretty standard stuff. The interviewer was easy to talk to and actually tailored the problem to my background, which made it feel less like a test and more like a working session. Coded slower than I'd like but got through it, and had a VO invite two days later.
  • Design and implement a versioned key-value store, framed around operations common in an ads data platform.

“The interviewer noticed I work in ads and suggested we reframe the problem around that context, which was a nice touch.”

View Post

Netflix·Software Engineer·Technical Phone Screen

Jun 2026
Netflix coding round for a software engineer position. The problem was a sliding window question but with a bitmask twist I wasn't fully expecting. Interesting experience overall, felt like they wanted to see if you understood the tradeoffs not just the solution.
  • Given an array of name strings, find the longest contiguous subarray where all names are unique. The constraint: you must implement it using a bitmask instead of a hash set. Be prepared to justify correctness and discuss complexity tradeoffs.

“I knew the sliding window approach cold from the classic version, so I felt fine until they said bitmask.”

View Post