TikTok Interview Questions — Page 15

TikTok·Software Engineer·Onsite - System Design / Architecture

May 2026

TikTok system design round focused almost entirely on Redis internals. Pretty deep dive, more than I expected for a single session.

Walk me through Redis's core data structures and their time and space complexity characteristics.
Compare RDB and AOF persistence in Redis. What are the durability, performance, and recovery trade-offs?
How do Redis eviction policies work, and what happens to keys with TTLs when the instance is under memory pressure?
Explain how Redis replication works, how Sentinel handles failover, and how Cluster handles sharding.
What are the limitations of Redis transactions and Lua scripting, and what are the common pitfalls when implementing distributed locks?
How would you prevent a cache stampede, and what strategies exist for handling hotspot keys in Redis?

“Covered strings, lists, sets, sorted sets, hashes, and streams.”

View Post

TikTok·Software Engineer·Onsite - Behavioral / Leadership

Jun 2026

Behavioral round at TikTok for a software engineer role, basically just one question that they really wanted you to unpack from every angle.

Tell me about the hardest technical or work problem you've dealt with recently. Walk through why it was hard, what solutions you considered, what you actually did, and what the results were. Also, what would you change about how you handled it?

“This is the kind of question that sounds manageable until you realize they want numbers.”

View Post

TikTok·Data Scientist·Onsite - Multi Round

Jun 2026

Three rounds for a Data Scientist role at TikTok, mixing ML theory, system design, and coding. The questions recycled a few themes across rounds (Dropout came up twice, which was a little odd) and the coding problems leaned more LeetCode-style than pure DS. Felt like a solid process but the repetition made me wonder if the rounds were coordinated at all.

How would you deploy multimodal models under tight compute and GPU memory constraints?
What is overfitting and what are some ways to mitigate it?
Implement a MinStack that returns the minimum element in O(1) time.
What methods help reduce overfitting in deep learning, and how does Dropout work at a principled level?
Compare different normalization techniques (batch norm, layer norm, etc.) and explain how each behaves differently at inference time.
How is reinforcement learning applied in LLM post-training, specifically in the context of learning from human feedback?
Implement a MaxStack. Then, extend the idea to compute a running median from a data stream.
Explain Dropout again and specifically address why it preserves distributional consistency between training and inference.
Given a binary tree, determine if there exists a path starting from any node and moving only upward toward the root where the node values sum to a given target.

“This one took me a second to find my footing.”

View Post

TikTok·Software Engineer·Onsite - System Design / Architecture

Jun 2026

System design round at TikTok for a software engineer role, and they did not ease you in gently. The whole session was essentially one massive distributed systems question with about six sub-problems layered inside it.

Design a globally distributed notification service that handles real-time and scheduled messages across email, SMS, and push channels for tens of millions of users, while respecting regional data compliance requirements.
How would you define the APIs and data models for this notification service?
How do you handle deduplication and idempotency in a high-throughput notification pipeline?
What rate limiting strategy would you apply across different notification channels and user tiers?
Walk me through your choice of storage and queuing layers for this system.
How would you orchestrate workers, handle retries with backoff, and preserve message ordering guarantees?
How would you design multi-region failover and disaster recovery for this notification service?
Provide rough capacity estimates and describe your monitoring and alerting strategy for this system.

“This is a beast of a question and I underestimated how much ground it covers.”

View Post

TikTok·Data Scientist·Technical Phone Screen

Jun 2026

TikTok data scientist interview with a SQL question around identifying repeat advertisers across consecutive weeks. Pretty focused and technical, nothing too wild.

Given a table of weekly advertiser ad submissions, write a SQL query to return all advertiser IDs that submitted ads in both Week 1 and Week 2.

“Went with a self-join first, which worked but felt clunky when I looked back at it.”

View Post

TikTok·Data Scientist·Onsite - Behavioral / Leadership

Jun 2026

TikTok data scientist interview with a pretty brutal project deep-dive question that covers basically everything at once. One question, eight parts, two to three minutes to answer it all. Not a vibe if you haven't pre-built a tight story.

Walk me through the single most impactful project you led end-to-end. Cover the problem and business context, your role and team, baseline vs. final metrics with real numbers, two approaches you rejected and why, the hardest trade-off you made and how you justified it, one major risk you de-risked and what would have changed if you were wrong, a conflict you navigated, and what you'd do differently if you ran it again next quarter.

“This is a monster of a question and I did not have it all loaded up cleanly.”

View Post

TikTok·Machine Learning Engineer·Technical Phone Screen

Jun 2026

TikTok ML Engineer interview with a classic sliding window problem. Nothing too exotic but the O(n) constraint means you can't just brute force it and hope for the best.

Given a string, find the length of the longest contiguous substring with no repeated characters.

“Sliding window with a hash set is the move here.”

View Post

TikTok·Machine Learning Engineer·Online Assessment (OA)

Jun 2026

TikTok ML engineer interview with a coding problem focused on finding the largest numbers. Not much else to go on from what I remember.

Given a dataset or array, find the largest numbers according to some criteria.

“Pretty bare bones problem on the surface.”

View Post

TikTok·Data Scientist·Technical Phone Screen

Jun 2026

SQL-heavy data science interview at TikTok, all four questions were about querying two streaming tables. Nothing behavioral, just back-to-back SQL with some tricky edge cases around window functions and joins.

Write a SQL query to calculate total streamed hours per month, including a solution that handles data spanning multiple years.
For each streamer, calculate their total streaming duration and the ratio of time spent in a specific category to their total duration. How do you handle queries when the category name is case-sensitive or passed as a keyword?
Identify streamers who streamed more in a given month than they did the previous month. How do you handle year boundaries and months with no data (NULL values)?
Review this SQL query that calculates average concurrent viewers per streamer in 2019 and total view time from US viewers. Is it correct, and if not, what's wrong with it?

“Pretty straightforward at first glance.”

View Post

TikTok·Product Manager·Onsite - Product Sense / Strategy

Jun 2026

TikTok PM interview that covered a mix of product sense and trust/safety thinking. Three questions, all pretty meaty, and the privacy one especially felt like it could go in a dozen directions.

Walk through TikTok's current login flow from a user perspective. What parts work well and what creates friction?
Choose one TikTok feature and propose meaningful improvements. What user problem are you solving, what design decisions are you making, and how would you measure success?
What safeguards would you put in place to protect user data privacy and maintain data integrity on the platform?

“I started by just narrating the flow step by step like a user would experience it, which helped me stay grounded.”

View Post