英伟达深度学习实习生面试指南

简历筛选

先把自己的简历做好，最好英文的，投递简历后一般要需要几天才知道简历筛选过了没有。

笔试

如果比较幸运，过几天会收到HR的电话，告诉你通过简历筛选，跟你商量笔试时间，一般两三道编程题，并针对此问题考察一些算法复杂度和时间复杂度的计算。笔试主要考察基本的编程能力和算法时间负责度、空间复杂度的知识。

电话面试

电话面试的面试小哥态度非常好，非常亲切，会先告诉你电话面试的基本安排，然后首先是跟你讲一讲部门team所从事的任务。我面试的是Deep Learning Compute Architect部门，给我介绍了其主要从事的任务。最后告诉我需要的Domain Knowlede主要分为三块：

1.Programming：1)C/C++/Python 2)Data structure 3)algorithm

2.计算机体系结构

3.性能分析+建模

职位要求

基本要求

严谨的逻辑思维和分析、有较强学习能力、熟悉深度学习算法实现和框架。

有较强计算机体系结构背景和编程能力。

有针对DNN算法设计加速硬件经验加分。

工作职责

跟踪学术和工业界最新研究成果、针对深度学习未来趋势分析和研究；

提出GPU未来发展的方向和重点。

针对DNN算法特征提出基于处理器的硬件加速方案。

What you’ll be doing:

1. Develop innovative HW, DSP, GPU and system architectures to extend the stat

e of the art in deep learning performance and efficiency

2. Analyze and prototype key deep learning and data analytics algorithms and a

pplications

3. Understand and analyze the interplay of hardware and software architectures

on future algorithms and applications

4. Collaborate across the company to guide the direction of machine learning,

working with software, research and product teams

What we need to see:

1. MS or equivalent experience

2. Track record of designing architectures to accelerate computational demandi

ng algorithms and applications

3. Strong mathematical foundation in machine learning and deep learning

4. Experience working with deep learning frameworks like Caffe, TensorFlow and

Torch

5. Strong programming skills in C, C , Perl, or Python

6. Familiarity with GPU computing (CUDA, OpenCL) and HPC (MPI, OpenMP)

7. Strong background in computer architecture

8. Experience with systems-level performance modeling, profiling, and analysis

9. Experience in characterizing and modeling system-level performance, executi

ng comparison studies, and documenting and publishing results.