← Back to Directory

LinkedIn interview notes, page 12 of 32

Large Enterprises

Most of these LinkedIn write-ups come from Software Engineers (156), then Product Managers (66) and Data Scientists (37). The most reported rounds are the technical phone screen (169), the product sense round (62), the system design round (42) and the behavioral / leadership round (20).

316 interview notes · updated Oct 2026

LinkedIn·Machine Learning Engineer·Technical Phone Screen

May 2026
LinkedIn ML Engineer interview that went deep on optimization theory. One question on Adam took up a big chunk of time and I wasn't fully prepared for how far they'd push on the intuition behind it.
  • Walk through how Adam works, including its first and second moment estimates, the bias correction step, and the per-parameter update rule. Then explain why dividing by the square root of the second moment adapts the learning rate per parameter, how that connects to RMSProp, and what it means for sparse or unevenly scaled gradients.

“I got the mechanics down fine, m_t and v_t, bias correction, the update formula.”

View Post

LinkedIn·Software Engineer·Onsite - System Design / Architecture

May 2026
LinkedIn system design screen for a software engineering role. The question was about building a Top-K search query tracker, which sounds straightforward until you actually try to scope it properly under time pressure.
  • Design a service that records user search queries and returns the Top-K most searched terms, with support for both all-time counts and a recent time window variant.

“Spent the first few minutes fumbling through clarifications I should've had a mental checklist for.”

View Post

LinkedIn·Software Engineer·Technical Phone Screen

May 2026
LinkedIn technical phone screen for a Software Engineer role, focused entirely on iOS memory management under ARC. Pretty deep dive for a single topic, they gave me a time box and just let me talk.
  • Walk me through how ARC manages object lifetime, including what happens at compile time and when objects actually get deallocated.
  • Explain strong, weak, and unowned references. When would you actually reach for each one?
  • Where do retain cycles commonly show up in iOS code, and how do you prevent or break them?
  • What is an autorelease pool and when would you actually need to use one manually in Swift?
  • How would you debug memory leaks or unexpectedly retained objects in Xcode?

“Started okay, said ARC inserts retain/release calls at compile time so you don't manage reference counts manually.”

View Post

LinkedIn·Software Engineer·Onsite - System Design / Architecture

May 2026
System design round at LinkedIn for a software engineer role. The whole session was one big question about building a multi-timezone calendar, which sounds contained but unravels fast once you get into recurring events and DST.
  • Design a calendar system that supports users across multiple time zones, including one-time and recurring events, schedule queries over a time range, correct timezone handling, and daylight saving time edge cases.

“I started with the API and data model which felt safe, but the recurring events piece is where I started to lose the thread.”

View Post

LinkedIn·Software Engineer·Technical Phone Screen

May 2026
LinkedIn follow-up round digging into a rank-based eviction cache I'd apparently designed in a previous session. Both questions were direct continuations of that design, so if you haven't thought through mutable signals in caching before, this will hurt.
  • If two cached entries have the same rank, how do you break the tie? Walk through using a last-seen timestamp as a secondary sort key and how a doubly-linked list in access order supports that.
  • The rank value used for eviction can change over time. What are the implications of building an eviction policy on a mutable signal, and how do you handle stale heap entries, lazy deletion, or periodic re-heapification versus just switching to a monotonic signal like access time?

“I'd thought about LRU before so the doubly-linked list part felt familiar, but connecting it explicitly to a secondary key in a rank-based system took me a second to articulate.”

View Post

LinkedIn·Product Manager·Onsite - Product Sense / Strategy

May 2026
LinkedIn product interview, one question about designing a guest experience for YouTube. Short and focused, basically a pure product sense exercise with no warm-up.
  • How would you design a guest experience for YouTube?

“I went straight to defining who the guest user actually is, someone not signed in, and tried to map out what they can and can't do versus a logged-in user.”

View Post

LinkedIn·Software Engineer·Technical Phone Screen

May 2026
LinkedIn SWE coding round, pretty focused on DP. One problem, one optimization discussion, and then it was over. Left feeling like I could've explained the O(n*k) trick more cleanly.
  • Given n houses in a row and k colors, where costs[i][j] is the cost to paint house i with color j and no two adjacent houses can share the same color, find the minimum total cost to paint all houses.

“I got the DP setup pretty fast.”

View Post

LinkedIn·Software Engineer·Technical Phone Screen

May 2026
LinkedIn coding round, looked like a straightforward frequency problem but the tie-breaking condition is where people trip up. Worth knowing your heap mechanics cold before walking in.
  • Given a list of words and an integer k, return the k most frequent words ordered by frequency descending, with ties broken alphabetically.

“The frequency counting part is fine, everyone gets that.”

View Post

LinkedIn·Software Engineer·Technical Phone Screen

May 2026
LinkedIn software engineering interview focused on a classic hard data structures problem. The LFU cache question was the whole session, and they wanted a full working implementation with O(1) guarantees, not just a sketch.
  • Design and implement an LFU cache supporting get and put operations, both in O(1) average time. On eviction, remove the least frequently used key, breaking ties by recency.

“The two-hashmap approach clicked for me eventually: one maps keys to nodes, the other maps frequency counts to a doubly linked list of nodes at that frequency.”

View Post

LinkedIn·Product Manager·Onsite - Product Sense / Strategy

May 2026
LinkedIn PM interview, one question about feature development and go-to-market. Pretty standard stuff but it's broader than it sounds and I didn't pace myself well.
  • You're building a new feature for a product. Walk through how you'd take it from idea to launch.

“I jumped straight into discovery and forgot to anchor on who the user was first.”

View Post