← Back to Directory

ByteDance interview notes, page 19 of 29

FAANG / Big Tech

Most of these ByteDance write-ups come from Software Engineers (194), then Machine Learning Engineers (45) and Data Scientists (17). The most reported rounds are the technical phone screen (227), the system design round (28), the behavioral / leadership round (16) and the onsite coding round (9).

289 interview notes · updated Sep 2026

ByteDance·Machine Learning Engineer·Technical Phone Screen

Jun 2026
Bytedance ML engineer interview with a classic dynamic programming problem. Nothing too exotic but the pressure of getting it right on the spot is real.
  • You have a row of houses, each with some amount of money. You can't rob two houses next to each other. What's the maximum you can rob?

“Classic DP, I knew it the second they described it.”

View Post

ByteDance·Software Engineer·Technical Phone Screen

Jun 2026
Bytedance software engineer interview with a brain teaser that sounds simple but has a neat trick to it. Not your typical coding round.
  • A worker is paid one gold bar for seven days of work. You can only make two cuts to the bar. How do you pay exactly 1/7 of the bar at the end of each day?

“I stared at this for a while.”

View Post

ByteDance·Backend Engineer·Technical Phone Screen

Jun 2026
Bytedance backend screen, just one question about OS fundamentals. Pretty short session, felt like a warmup but they seemed to care about the answer more than I expected.
  • What is the difference between a thread and a process?

“I knew this one but fumbled the memory model part.”

View Post

ByteDance·Software Engineer·Technical Phone Screen

Jun 2026
Bytedance SWE interview, got a classic interval problem that looks straightforward until you're actually coding it under pressure. Nothing too exotic but the details matter.
  • Given a sorted list of non-overlapping intervals and a new interval, insert the new interval and merge any overlaps. Return the result sorted and non-overlapping.

“I knew the general shape of the solution pretty fast: walk through the list, dump everything that ends before the new interval starts, then keep merging anything that overlaps by stretching the new interval's bounds, then tack on the rest.”

View Post

ByteDance·Software Engineer·Technical Phone Screen

Jun 2026
Bytedance SWE coding round, roughly an hour. The problem was a hard-level LeetCode variant and I didn't finish cleanly, but somehow passed anyway.
  • Implement an LRU Cache with TTL support, where each entry can have its own expiration time passed at call time. Handle expired entries correctly across get and put operations.

“This is already a hard problem on its own and adding TTL made it significantly more involved.”

View Post

ByteDance·Machine Learning Engineer·Technical Phone Screen

Jun 2026
Bytedance ML engineer interview that went deep on loss functions, way deeper than I expected for a phone screen. The KL divergence part especially had some follow-ups that made me realize I only half-understood what I thought I knew.
  • Walk through the common loss functions (MSE, MAE, Huber, cross-entropy, hinge, focal) and explain when you'd pick each one.
  • Define KL divergence, explain its asymmetry, and describe how it relates to cross-entropy. When would you use it over a simpler loss?

“Started fine with MSE vs MAE, outlier sensitivity, the usual.”

View Post

ByteDance·Software Engineer·Technical Phone Screen

Jun 2026
Bytedance SWE interview that centered on a stream processing problem with a validation twist. The coding part was straightforward enough but the follow-up discussion about batching vs in-memory caught me a bit off guard.
  • Given a stream of records each with a timestamp, date, and username, count how many times each user logged in per day. Output should be sorted by username then by day. Also validate the input records and skip or flag any with malformed dates, invalid times, or mismatches between the timestamp and date fields.
  • How would you handle this same problem if the records are arriving as a live stream versus if you have all the data in memory at once? What changes in your approach?

“I got the aggregation part pretty fast, basically a hashmap keyed on (user, day) and then sort at the end.”

View Post

ByteDance·Software Engineer·Technical Phone Screen

Jun 2026
Bytedance coding interview for a Software Engineer role. One algorithmic question that I couldn't fully crack, and I left feeling pretty bad about it. Also worth knowing: they've moved to Feishu for everything now, including live coding.
  • Given an N-ary tree, find the number of paths that sum to a given target value. The path does not need to start or end at the root or a leaf.

“This is basically LeetCode 437 but with an N-ary tree instead of a binary one.”

View Post

ByteDance·Software Engineer·Technical Phone Screen

Jun 2026
Bytedance SWE coding round, one algorithmic problem on subsets with duplicates. Pretty standard backtracking territory but the dedup logic is the part that actually matters and where I had to think carefully.
  • Given an integer array that may contain duplicates, return all possible unique subsets (the power set) without any duplicate subsets.

“I knew the no-duplicates version cold, so my first instinct was to just write that and slap a set on the output.”

View Post

ByteDance·Software Engineer·Technical Phone Screen

Jun 2026
Bytedance software engineer round, one coding question on expression evaluation. Pretty standard stuff but the implementation details trip you up if you haven't seen it before.
  • Implement a basic calculator that evaluates a string expression containing non-negative integers, the four arithmetic operators, and spaces, without using any built-in eval functions.

“I knew the stack approach going in but kept second-guessing myself on when to apply multiplication and division versus just pushing things.”

View Post