Dynamic Programming
Twitch runs its coding interviews on Amazon's rails, and dynamic programming sits on those rails as a breadth check rather than a signature. It shows up to confirm you can build a recurrence from scratch, not to test some streaming-specific trick. By weight it is one of the smaller families in the section, behind the arrays and the data-structure design work that open most loops. What makes it worth real preparation is the tail. When a round does spike to a hard problem, it usually spikes here. Concatenated Words and Binary Tree Maximum Path Sum are the ones candidates remember, and both live in this territory.
The other thing to know going in is that DP arrives without product framing. The rest of the coding section borrows shape from Twitch's own systems. A trees-and-graphs round can lean on the fact that a chat room's moderation rules or a category tree are genuinely tree-shaped. The heaps round poses its top-K as a ranking of viewers by watch time. Dynamic programming has no equivalent in the product. The genuinely hard optimization problems at Twitch, choosing a bitrate ladder for a transcode or reconciling message order across a chat cluster, live in the system-design rounds, not in the coding hour. When DP appears in the coding hour, it appears as a canonical string problem with the plain structure showing.
- 3 problems, worked step by step
- about 24 min
Premium lessonTwitch interview guide
These took digging.
What candidates walked out with, from 20,000+ reports, forum threads behind logins and write-ups in Chinese. Then solved step by step.
100 companies. Checked every month.
Unlock all 100 companies$49 a year · less than $1 a week · 30-day money-back guarantee
Twitch interview guideKnow what Twitch asks before you walk in.
14 coding · 5 system design · 39 behavioral questions
Get their questions first.
InterviewLegend is independent and not affiliated with or endorsed by any company it covers.