How Nvidia Interviews in 2026: Process, Questions & What They Score

1. 01

   Print numbers 1 to n without using any loop construct (recursion-based), a specifically and repeatedly reported Nvidia question.

   Tests recursion as an alternative to iteration; a frequently reported Nvidia prompt for printing 1 to n without loops.

2. 02

   Check whether one string exists as a substring within another.

   Tests substring search; be ready to discuss the naive approach vs KMP-style linear matching.

3. 03

   Reverse a linked list, both iteratively and recursively.

   Tests linked-list pointer manipulation, both iteratively and recursively.

4. 04

   Number of islands / connected-components style grid traversal problem.

   Tests grid BFS/DFS for connected components, the classic number-of-islands pattern.

5. 05

   Detect a cycle in a linked list.

   Tests cycle detection in a linked list; the expected insight is Floyd's fast and slow pointers.

6. 06

   Problems emphasizing performance-conscious thinking: memory locality and cache-friendly access patterns, prefix sums/scans, and bitwise operations.

   Tests low-level performance thinking: cache-friendly memory access, prefix sums, and bitwise operations.

7. 07

   General LeetCode-medium-difficulty problems on arrays, strings, trees, and graphs, with follow-up questions on time/space complexity and edge cases.

   LeetCode-medium range across arrays, strings, trees, and graphs, with follow-ups on complexity and edge cases.

Reported problems, listed so you know what to expect. Practice them in your own editor or on the platform you prefer; SupaCV's practice mode coaches how you talk through them.

1. 01

   Design a ride-hailing platform like Uber.

   System design: they probe matching riders to drivers, real-time location updates, and dispatch at scale.

2. 02

   Design a proximity/location-based server.

   System design: tests spatial indexing (geohash or quadtree) and low-latency nearest-neighbor queries.

3. 03

   Design a cloud-based file storage system.

   System design: they probe chunking, metadata services, deduplication, and sync for large-scale file storage.

4. 04

   Design a job scheduler or telemetry pipeline for a GPU cluster.

   System design: tests job scheduling and telemetry collection for GPU clusters, including queueing and resource allocation.

5. 05

   Design a distributed inference system meeting a target request rate and latency across many GPUs; expect follow-ups on model/tensor vs. pipeline parallelism, interconnects (e.g., NVLink/InfiniBand), and fault tolerance ('what happens if a node crashes?'). Note: the first few items in this list are generic system-design prompts seen across many companies' prep guides, not Nvidia-exclusive; the GPU/inference-flavored ones are the most Nvidia-specific and worth prioritizing.

   System design: distributed inference under latency targets; follow-ups probe parallelism strategies and node failures.

6. 06

   General trade-off discussions: sharding, scheduling, and fault-tolerant coordination for large-scale, high-throughput systems.

   Tests tradeoff reasoning on sharding, scheduling, and fault-tolerant coordination in high-throughput systems.

Reported problems, listed so you know what to expect. Practice them in your own editor or on the platform you prefer; SupaCV's practice mode coaches how you talk through them.

1. 01

   How do convolutional neural networks (CNNs) differ from traditional neural networks in processing image data?

   ML fundamentals: how convolution layers exploit spatial structure compared with fully connected networks.

2. 02

   Explain how Transformers work and where they're applied in NLP.

   Tests understanding of Transformer architecture, attention, and NLP applications.

3. 03

   What is the difference between Layer Normalization and Batch Normalization, and when would you use each?

   Tests normalization knowledge: LayerNorm vs BatchNorm behavior and when each fits.

4. 04

   Explain log loss and its significance as a model evaluation metric.

   Tests understanding of log loss as an evaluation metric and its penalty on confident wrong predictions.

5. 05

   How would you account for variation between two groups when analyzing an A/B test?

   Tests A/B testing statistics: controlling for between-group variation when analyzing results.

6. 06

   Explain overfitting and underfitting and how you would address each.

   Tests diagnosing overfitting vs underfitting and remedies such as regularization or added model capacity.

7. 07

   Write a SQL query to summarize revenue/other metrics from a dataset.

   SQL exercise; tests aggregation and grouping to summarize revenue or similar metrics.

Reported problems, listed so you know what to expect. Practice them in your own editor or on the platform you prefer; SupaCV's practice mode coaches how you talk through them.

1. 01

   Explain memory coalescing in CUDA and its impact on performance.

   CUDA fundamentals: tests understanding of memory coalescing and why access patterns drive kernel performance.

2. 02

   What is warp divergence, and how do you minimize it?

   Tests understanding of warp divergence and branch restructuring techniques to minimize it.

3. 03

   What is the difference between CUDA cores and Tensor cores?

   GPU architecture knowledge check: CUDA cores vs Tensor cores and the workloads each serves.

4. 04

   How would you use Nsight Systems vs. Nsight Compute to profile and optimize a kernel?

   Tests practical profiling workflow: when to use Nsight Systems vs Nsight Compute to find kernel bottlenecks.

5. 05

   Explain cache design/performance, virtual-to-physical address translation, and the role of the TLB.

   Computer architecture fundamentals: cache behavior, virtual-to-physical address translation, and the TLB's role.

6. 06

   Explain pipelining in a processor and its performance implications.

   Tests processor pipelining concepts: stages, hazards, and performance implications.

7. 07

   (ASIC/verification-track roles) UVM basics and testbench development; a specifically documented Nvidia ASIC verification question: if the design engineer and verification engineer both make the same mistake in the testbench BFM and RTL/DUT, how do you still detect the error (answer hinges on independent spec-based checks, not just reference-model comparison)?

   Verification-track UVM question; the insight is independent spec-based checks that catch correlated DUT and testbench errors.

Reported problems, listed so you know what to expect. Practice them in your own editor or on the platform you prefer; SupaCV's practice mode coaches how you talk through them.

1. 01

   Why do you want to work at Nvidia?

2. 02

   Describe a time you disagreed with a senior engineer's technical approach. How did you handle it, and what happened?

3. 03

   Tell me about a time you received negative feedback and how you handled it.

4. 04

   Tell me about a time you had a conflict with a colleague. How did you resolve it, and what did you learn?

5. 05

   Describe a problem you identified that others on your team had missed.

6. 06

   Tell me about an experiment or project that failed. What did you learn from it? (probes 'Intellectual Honesty')

7. 07

   Describe a time you had to move quickly or adapt your plan on incomplete information. (probes 'Speed and Agility')

8. 08

   Describe a time you collaborated across teams on a project. What was your role, and how did you help the group succeed? (probes 'One Team')