← Back to Directory

NVIDIA interview notes, page 5 of 17

FAANG / Big Tech

Most of these NVIDIA write-ups come from Software Engineers (137), then Product Managers (15) and Data Scientists (10). The most reported rounds are the technical phone screen (106), the system design round (28), the hiring manager screen (13) and the product sense round (13).

166 interview notes · updated Aug 2026

NVIDIA·Software Engineer·Onsite - System Design / Architecture

Apr 2026
Interviewed for a solutions architect role at Nvidia. Two meaty system design questions back to back, both in the distributed computing space. Felt like they were probing for depth rather than breadth.
  • Walk through how you would design a distributed training system that scales across hundreds of nodes.
  • How do you handle node failures in a large-scale distributed system?

“This one took up most of the time.”

View Post

NVIDIA·Software Engineer·Technical Phone Screen

May 2026
Prepped for an NVIDIA software engineer role and the first thing they hit me with was a self-intro. Felt like a warm-up but it set the tone for everything after.
  • Give a 2-3 minute introduction of yourself: your current role, technical background, a recent project with measurable impact, and why you're interested in this company and role specifically.

“I over-rehearsed this and it showed.”

View Post

NVIDIA·Software Engineer·Technical Phone Screen

May 2026
Nvidia SWE interview with a pretty focused question on array manipulation. Not a lot to go on in terms of rounds or outcome, but the question itself had some nuance worth thinking through.
  • You're given multiple arrays of different lengths. How do you pad them efficiently?

“Seems basic until you start thinking about what 'efficient' actually means here.”

View Post

NVIDIA·Software Engineer·Technical Phone Screen

May 2026
NVIDIA SWE interview that went deep on ML framework internals. The question was open-ended but don't let that fool you, they clearly expected you to have real opinions and not just recite docs.
  • What are the major high-level trends happening at the deep learning and machine learning framework level right now, and how do PyTorch and JAX compare across their programming models, compilation strategies, and ecosystem support for distributed training and hardware portability?

“This one sprawled fast.”

View Post

NVIDIA·Software Engineer·Technical Phone Screen

May 2026
NVIDIA software engineer interview that centered almost entirely on a run-length encoding problem. Pretty implementation-heavy for what I expected to be a more algorithmic round, but the follow-up pushed into real depth fast.
  • Implement a compress function that encodes a string using run-length encoding, where every character run is represented as the character followed by its count (including runs of length 1).
  • Implement the inverse decompress function that takes a run-length encoded string with potentially multi-digit counts and reconstructs the original string.
  • How would you verify the round-trip property (compress then decompress returns the original, and vice versa), and what is the time and space complexity of each function?
  • Extend the decoding to support nested bracket notation like '3[a2[bc]]', where a number before brackets means repeat the enclosed string that many times.

“Felt straightforward at first and I almost rushed into code without thinking about the single-character case.”

View Post

NVIDIA·Software Engineer·Onsite - System Design / Architecture

May 2026
System design round at NVIDIA for a software engineering role. The question was a single sprawling design problem that covered a lot of ground, and I felt like I kept getting pulled in different directions trying to cover everything.
  • Design a bidirectional data synchronization platform that ingests planned maintenance and outage events from cloud providers like AWS, GCP, and Azure, normalizes them into a shared schema, and surfaces them on an internal dashboard. You should also cover how updates flow back to providers, how you ensure durability and idempotency, how fresh the dashboard stays, and what your observability story looks like.

“This one is massive.”

View Post

NVIDIA·Software Engineer·Technical Phone Screen

May 2026
NVIDIA coding round, one design question the whole time. Pretty focused on getting the implementation right rather than just talking through ideas.
  • Design and implement a disk space manager with a fixed total capacity. It should support storing datasets by ID and size, retrieving a dataset's size, and evicting existing datasets when space is needed. The capacity constraint is on total bytes used, not on the number of items stored.

“The tricky part I kept second-guessing was the capacity semantics.”

View Post

NVIDIA·Data Scientist·Technical Phone Screen

May 2026
Technical screen for a Data Scientist role at NVIDIA, heavily GPU-focused. The whole thing was basically a deep dive into CUDA internals and GEMM kernel optimization, which felt more like a GPU systems engineer interview than anything data science adjacent.
  • Walk through CUDA's execution model: grids, blocks, warps, and threads. How do they relate to each other and to the hardware?
  • Describe CUDA's memory hierarchy: global, shared, registers, constant, and texture memory. What are the performance implications of each?
  • What is occupancy in CUDA and how do you reason about it when designing a kernel?
  • Design a naive single-precision GEMM kernel for C = A x B, then walk through how you would optimize it step by step.
  • How do you choose grid and block dimensions for a GEMM kernel, and how do you handle matrix sizes that aren't multiples of your tile size?
  • How would you compare your custom GEMM kernel's throughput against cuBLAS?
  • How would you overlap host-to-device memory transfers with GPU compute using CUDA streams and pinned memory?
  • How do you measure achieved occupancy, arithmetic intensity, and memory bandwidth for a CUDA kernel?

“I had this down pretty well.”

View Post

NVIDIA·Software Engineer·Onsite - Product Sense / Strategy

May 2026
Got a product strategy question at NVIDIA for a software engineer role that felt more PM-adjacent than I expected. Just the one question but it made me think harder than any coding problem would have.
  • Beyond performance benchmarks, what other content pillars would you propose for a software product's blog? For example, how would you think about covering things like observability features or reducing the learning curve for new users?

“I defaulted to metrics and benchmarks at first because that's what most devs lead with, then caught myself and tried to reframe around what actually builds trust with a technical audience over time.”

View Post

NVIDIA·Software Engineer·Technical Phone Screen

May 2026
NVIDIA software engineer interview that leaned hard into algorithmic problem solving. The sliding window section had two back-to-back problems with a complexity analysis follow-up, which felt like a lot to cover in one sitting.
  • Given an array of positive integers and a target sum S, find the minimal length of a contiguous subarray whose sum is at least S. Return 0 if no such subarray exists.
  • Given a string and an integer K, return the length of the longest substring that contains at most K distinct characters.
  • Analyze the time and space complexity of your sliding window solutions, and explain when you'd choose a fixed-size window versus a dynamic one.

“I knew this was a dynamic window problem pretty quickly but fumbled the shrinking logic at first.”

View Post