Figma Interview Questions — Page 2

Figma·Software Engineer·Onsite - System Design / Architecture

May 2026

Figma system design round focused entirely on a document-editing layer with undo/redo semantics. The problem kept expanding, which I wasn't fully ready for, but it was a genuinely interesting problem to work through.

Design a document-editing layer that supports applying edits and undoing them. Implement an API with apply(Operation op), undo(), and getText().
Extend the design to support transactional batching: beginBatch(), apply(...), commitBatch(), where undo() must revert an entire committed batch as a single atomic unit.
How would you optimize batch undo for very large batches containing millions of small operations, in terms of both time and memory?
Implement redo() that works correctly for both single edits and committed batches, and explain how new edits should invalidate the redo stack.
Describe how you'd handle edge cases like overlapping range edits and cursor bookkeeping within this editing layer.

“Started with a stack-based approach, two stacks, one for history and one for redo.”

View Post

Figma·Software Engineer·Technical Phone Screen

Jun 2026

Figma SWE interview that went deep on a layered drawing editor with undo/redo. The design side took longer than I expected and the edge cases around redo invalidation tripped me up a bit.

Design and implement a layered image editor that supports adding, removing, reordering, and editing layers, plus undo and redo. How do you handle edge cases like undoing when the stack is empty, or a new operation wiping the redo history?

“The undo/redo skeleton felt familiar but I fumbled the redo invalidation rule.”

View Post

Figma·Software Engineer·Onsite - System Design / Architecture

Jun 2026

Figma infra team interview, basically one long system design session about running a monolith on Kubernetes. Covered a lot of ground, from deployment topology all the way to disaster recovery and migration strategy. Dense question, and I think I undercooked the migration path section.

You're joining an infra team running a single monolithic backend on Kubernetes, with no plans for microservices or aggressive scaling. How would you design and operate it? Walk through deployment topology, networking, config and secret management, CI/CD, and rollout strategies.
How would you handle resource requests and limits, autoscaling, and stateful dependencies like databases, caches, and object storage in this Kubernetes setup?
What does observability look like for this monolith, and how would you handle incident response and disaster recovery?
What are the real trade-offs of staying monolithic versus moving to microservices in this context, and what concrete signals would tell you it's time to evolve the architecture?
If those signals do appear, how would you outline a low-risk incremental migration away from the monolith? Think about service boundaries, data ownership, API contracts, and how team workflows change.

“I started with pod replicas and node pools, talked about using a Deployment with a reasonable replica count and keeping stateless app nodes separate from anything touching persistent workloads.”

View Post

Figma·Software Engineer·Technical Phone Screen

Jun 2026

Figma coding interview with a two-part problem centered on resizing rectangles. The second part added a spatial constraint that made the first part feel like a warmup in hindsight.

You have a list of rectangles each with a height (minimum height is 1). Given a resize value, distribute it evenly across all rectangles. If positive, each gets resize/count added. If negative, reduce as evenly as possible without letting any rectangle drop below height 1. For example: resize=-12 with heights [3,5,10,9] should produce [1,1,7,6].
Extend the previous problem: each rectangle now also has a y-coordinate. After resizing, the relative positions between rectangles must be preserved. Note that relative position here is not just the difference in y-values but needs to account for the heights of the rectangles themselves.

“The positive case felt trivial and I almost rushed past it, which was a mistake because the negative case is where the real logic lives.”

View Post

Figma·Data Scientist·Recruiter / HR Screen

Jun 2026

Got a Figma DS offer and had to work through the full compensation math before deciding how to respond. The recruiter gave a range upfront which was helpful, but there were enough moving parts (RSUs, refreshers, vesting structure) that it took a while to figure out what I actually wanted to push on.

Given a base range of $210-225k and $600k in RSUs vesting over 4 years, walk through how you'd estimate your first-year total comp under different stock price scenarios (down 30%, flat, up 30%), then build a concrete counteroffer across base, equity grant, sign-on, and level.
What clarifications would you request before finalizing your response to the offer? Think about vesting structure, performance multipliers, promotion timelines, location banding, and start-date flexibility.
If you had to accept a trade-off between base salary and RSU grant size, what specific exchange would you be willing to make, and why?

“This is the kind of question that sounds like a math problem but is really about knowing your own priorities.”

View Post

Figma·Machine Learning Engineer·Onsite - System Design / Architecture

Jun 2026

Figma ML engineer interview that was more systems-heavy than I expected. The whole session centered on one design problem about undo behavior in a layer editor, and they kept pushing on edge cases I hadn't thought through.

Design a batch operation primitive for a layer/document editor where multiple mutations can be grouped and treated as a single atomic unit for undo. How do you structure the operation log, handle nested batches, and deal with partial failures mid-batch?

“Started with the basic command pattern and felt pretty good about it, but then they asked about nested batches and I kind of stalled.”

View Post

Figma·Software Engineer·Technical Phone Screen

Jun 2026

Figma coding screen, one question the whole time. It was a custom data structure problem that looked straightforward but had a specific memory constraint that I almost missed entirely.

You're given a Document Layer object with apply(key, value) and get(key) methods. Add batch() to mark the start of a tracked batch, and undo() to revert all changes made since the last batch() call. The catch: you can only store each key's original pre-batch value once, no matter how many times it gets updated within the batch.

“The memory constraint is where I almost tripped up.”

View Post

Figma·Software Engineer·Onsite - System Design / Architecture

Jun 2026

Figma SWE system design round, focused on building an async job scheduler. The interviewer was pretty relaxed and the conversation stayed mostly on fault tolerance and correctness guarantees. Not a brutal round by any means, felt manageable if you'd seen similar problems before.

Design an async job scheduler that supports two modes: jobs that run immediately, and jobs that run after a specified delay (e.g. after 1 hour).
How do you guarantee exactly-once execution for jobs in a distributed scheduler?
What happens when a worker node fails partway through executing a job?

“The core design wasn't too bad.”

View Post

Figma·Software Engineer·Technical Phone Screen

Jun 2026

Figma data science interview that centered around a project presentation with a business application follow-up. Pretty straightforward format but the second question is where it gets real.

Walk us through a past project you've worked on using slides.
How could the project you just presented be applied to Figma, and what business impact do you think it could have?

“Picked a project I knew cold, which helped.”

View Post

Figma·Software Engineer·Onsite - System Design / Architecture

Jun 2026

Figma system design round, one big open-ended question about building a trending articles feed at scale. The scope was massive and I felt like I was chasing my own tail for most of it.

Design a system that surfaces trending articles in near real time for a large-scale product. Walk through your requirements, architecture, and key trade-offs.

“This question has a lot of surface area and I underestimated how much of it I'd need to cover.”

View Post