Derived from 1 submission for ALL. Format details may vary across different teams or locations.
The Nvidia intern interview process is rigorous and often includes unexpected hardware-focused questions, such as CUDA, GPU cores, and multithreading in GPU threads. Candidates should prepare for a mix of software (C/C++, Python, OS concepts) and hardware-related topics, despite the role being labeled as an internship.
Format: General technical questions covering programming fundamentals and past projects. Topics:
Exact Questions Asked:
InterviewDigeststruct in C++?Format: Unexpected shift to compiler design and byte-code parsing, with potential technical difficulties. Topics:
Exact Questions Asked:
Format: Problem-solving with a focus on memory allocation and pointer manipulation. Topics:
malloc vs calloc)Exact Questions Asked:
malloc/calloc).Format: Focus on GPU architecture, multithreading, and CUDA. Topics:
Exact Questions Asked:
The interview experience at NVIDIA was challenging due to unexpected hardware-focused questions (e.g., CUDA, GPU cores) and technical difficulties during the session. The candidate managed to answer most questions but found the shift in focus from software development to hardware-related topics surprising for an internship role.
The interview began with a brief introduction followed by technical questions related to Python, C/C++, and previous projects. Topics included differences between C++14 and C++17, utility functions, struct memory allocation, and comparisons between Python, C, and C++.
Differences between C++14 and C++17
Utility Function
Struct Memory Allocation
Python vs C vs C++
Previous internship experience
The discussion unexpectedly shifted to compiler design, including questions on byte-code parsing. Technical difficulties with one interviewer's audio and connection exacerbated the challenges.
Byte-Code Parsing
A coding problem was presented involving memory allocation for pixels in a matrix representing an image, utilizing malloc and calloc. Two pointer-related output-based questions were also asked.
Memory allocation for pixels in a matrix using malloc/calloc
Pointers-related output-based questions
The interview concluded with questions on operating system concepts, particularly multithreading in GPU threads, kernel-related topics, GPU cores, and CUDA.
Multithreading in GPU threads
Kernel-related questions
GPU cores
CUDA