[AMD Official Use Only - AMD Internal Distribution Only]
JD-1
WHAT YOU WILL BE DOING:
We are seeking highly motivated Software Engineering intern/co-op to join our team. In this role
You will develop a deep understanding of complex multi-core firmware solutions for
SmartNICs providing virtualized storage and networking
We will provide opportunities to
design, develop and validate control and datapath components
develop/ enhance test frameworks and test suites
Unit, Integration, System testing with automation test suite
Thorough understanding and custom drivers implementation for various AMD's chip
You will learn tools and processes used for developing large embedded firmware/
software
WHO WE ARE LOOKING FOR:
Programming in C/ C++ and Python
Knowledge of core computer science concepts such as object-oriented design, algorithm
design, data structures, problem soloving and complexity analysis
Experience in low-level programming, device drivers would be preferable
Experience in using Automation frameworks such as Pytest, PyATS, etc.
Knowledge of PCIe, NVME, Ethernet
Good interpersonal and commuications skill
Good team player
Excellent written and verbal communication skills
JD-2
Overview
Development of Diagnostics suites for test and validation of AMD’s hardware IP blocks.
Execution of Diagnostic suites for production system level test, system product validation as
well as for customer production test in the lead up to the success of AMD’s product launch into
market.
Key Responsibilities of Co-op/ Intern:
Responsible for execution of AMD Diagnostics suites and/or Diagnostics test codes that
may include code development in object-oriented programs and/or test suites building,
testing, and debugging for production.
Participate in AMD’s next generation Bring-up and enable Diagnostics tests for
production SLT (system level test) and ASIC qualification suites.
Work with Design, IP owners, Platforms and Product Engineering on design functional
and performance verification as well as production test and characterizations
requirements.
Debug of Diagnostics issue working closely with internal and external AMD teams.
Involve in the automating Diagnostics Suite release process.
Works on related projects and/or assignments as required to help with improving
productivity and efficiency.
What can you expect to learn?
Overview of AMD processor architecture design & technology.
Gain hands-on experience in development & debug of Diagnostics suites/ IP test codes.
[AMD Official Use Only - AMD Internal Distribution Only]
Learn the know-how and gain insights into How to Test all the advanced IP blocks in
CPU at system level of a complex and advanced semiconductor IC product.
Preferred experience
C/C++ programming experience
Familiarity with Unix/Linux internals
Basic knowledge in scripting languages, such as UNIX shell scripting or Ruby
Experience with software version control systems (e.g., GIT, Subversion)
Effective communication, Ability to work with minimal input and directions, Proactiveness
& Positive attitude.
Understanding of x86 architecture (or any Processor architecture)
Experience with Assembly language programming
Linux application or device driver development
Experience in Processor Validation, development of low-level software
Academic credentials
Bachelor's or Master’s in Electrical and Electronics Engineering or Computer
Engineering/ Science
JD-3
WHAT YOU’LL BE DOING:
Candidate has opportunitites to work on functional and performance analysis of
o
various enterprise tools and also research on LLVM compiler support for various
cutting edde modern programming lanugages. If good in LLVM compiler tool
chain and compiler internals, has potential opportunity to get hands on modern
programming standards. Candidate can leverage opportunity to understand end
to end flow of compiler development
o Opportunity to contribute and claim to the success of AMD CPU compilers
working with the world class compiler engineers and performance engineers in
AMD and in opensource communities
o Daily responsibilities may include root cause analysis of compiler issues,
communicating with the compiler engineers, participating in team meetings,
downstreaming LLVM community code, research on compiler functions across
AOCC and competition etc..
o By end of the COOP candidate is expected to have gained a good knowledge
and experience on end to end work flow of compiler development and release. A
good knowledge on functional support of LLVM, GCC and intel compilers.
Potential to gain confidence on compiler engineering with hands on to compiler
code base
KEY QUALIFICATIONS:
o Compiler design, LLVM/GCC compiler tool chain, Strong in C++. Any knowledge
in devops tools and process including gtihup, gerrit and Jenkins is an added
advantage
o Do not include specific years of experience, or use the word REQUIRED (use
preferred or desired).
WHAT WOULD SET YOU APART:
o Good problem solving skills. Appetite to learn and explore the unknown
WHO WE’RE LOOKING FOR:
o Bachelors in computer engineering from a reputed institute with good
communication skills who is looking for COOP in his/her final year
[AMD Official Use Only - AMD Internal Distribution Only]
JD-4
Responsibilities:
Model Optimization and Profiling:
o Analyze and profile the performance of various Deep Learning models,
identifying bottlenecks and areas for improvement specifically for CPU execution.
o Implement optimization techniques such as model quantization, pruning, kernel
fusion, and algorithmic modifications to enhance model efficiency on CPU
platforms.
o Optimize model weights and activations through quantization to enable efficient
fixed-point arithmetic execution on CPUs, leading to reduced memory footprint
and faster computation.
o Design and implement efficient data structures and algorithms to support model
optimization and execution.
CPU-Specific Optimization:
o Leverage CPU-specific instruction sets (e.g., AVX, SSE) and optimization
libraries (e.g., AMD BLIS, oneDNN) to maximize performance.
o Explore and implement multi-threading and parallelization strategies to effectively
utilize multi-core CPU architectures.
Collaboration and Communication:
o Collaborate closely with Deep Learning researchers, software engineers, and
hardware engineers.
o Communicate technical concepts and findings effectively to both technical and
non-technical audiences.
Required Qualifications:
Strong proficiency in programming languages C++ and Python.
Solid understanding of Data Structures and Algorithms (DSA) principles and their
application in software development.
Basic understanding of Deep Learning models, including convolutional neural networks
(CNNs), recurrent neural networks (RNNs), and transformer models, and their
fundamental concepts.
Excellent problem-solving and analytical skills.
Ability to work independently and as part of a team.
WHO WE’RE LOOKING FOR:
MTech/PhD in CSE
