GPU HPC Application Developer and Consultant

Location: Munich, Bavaria, DE

Company: Advanced Micro Devices

Apply now

Apply for Job

What you do at AMD changes everything 

At AMD, we push the boundaries of what is possible.  We believe in changing the world for the better by driving innovation in high-performance computing, graphics, and visualization technologies – building blocks for gaming, immersive platforms, and the data center. 

Developing great technology takes more than talent: it takes amazing people who understand collaboration, respect, and who will go the “extra mile” to achieve unthinkable results.  It takes people who have the passion and desire to disrupt the status quo, push boundaries, deliver innovation, and change the world.   If you have this type of passion, we invite you to take a look at the opportunities available to come join our team.


Job Description

The HPC Centre of Excellence (HPC COE) within AMD is growing and now seeks an experienced and enthusiastic GPU HPC applications developer based within Europe (ideally Germany). As a member of the HPC COE you will be a point of HPC focus and leadership within AMD, providing application performance and systems expertise and guidance both to our internal teams and externally to our partners and customers.


This role is based in 2 parts:


1) Starting Spring 2020, until May 2021 you will primarily be working with a Climate customer porting and developing their ICON code from OpenACC on Nvidia to an appropriate paradigm on AMD Radeon Mi-architecture. This part of the role will require working with the customer, external developers and winning partner in establishing the optimised architecture for phase-2 of the system at the Climate customers site. During this time you will also be involved in helping secure wins and providing consultancy for other GPU-HPC related opportunities travel with be about 10 - 15% with periodic travel to the customer site in Germany and a developer in Switzerland.


2) After May 2021. Provide support to general sales cycle by working with customers and partners and performing Proofs of Concept as required. Travel will increase to about 25% during this phase


In both phases there will be international travel, including the USA and possibly India.


This is a ‘hands-on’ developer role and requires an individual who can work independently to solve technical issues, but who can also reach into the other engineering teams across the business. Within AMD you will work with our customer-facing Field Application Engineering and

Sales teams, HPC Working Group, and other key engineering groups including those leading the CPU and GPU development, and compiler and math library teams.


Whilst this role requires technical engineering expertise, the person for this role will be required to lead technical meetings and own technical relationships with both customers and HPC partners. You will have the opportunity to work across a variety of customers on the design and performance of the worlds largest HPC systems and have a significant impact at a global level on the largest HPC systems.


Whilst we are seeking strong HPC application expertise we are also looking for someone who can grow in the role and become a senior member of engineering staff. AMD provides a career path for technical staff, either as an individual contributor up to AMD Fellow level, or as a future technical manager.


This is a key HPC role with a high degree of visibility both within and external to AMD. You will play a key part in growing the success and market penetration of the AMD EPYC CPU and GPU. If you are enthusiastic and passionate, looking for a deeper technical exposure of the CPU and GPU as they apply to HPC, want to be part of the great but still growing story of AMDs success in HPC, have confidence in an established career path, and are excited by the opportunity to influence and lead the largest global HPC opportunities then we are extremely keen to hear from you!


Role of the GPU Applications Expert

· Porting HPC codes to the AMD Radeon Mi-architecture from either CPU or other GPU architectures. This role will initially be focused on porting the ICON climate code currently written in OpenACC until May 2021

· In addition to development, to completely own and lead the technical engagement, lead technical meetings, set milestones and deadlines, communicate effectively between partner and customer, and within AMD. Periodic visits to the customer and communicating project milestones and outcomes at appropriate junctures.

· Collaborating with other ICON developers and driving development for the AMD platform

· Through constructive feedback, bug reporting and documentation across technical teams within AMD, the candidate will also shape product development and priorities

· Providing some support to our Field Application Engineers and Sales teams during the sales cycle where the customer may require a deeper level of expertise during the phase until May 2021; and then on a full-time wider sales engagement after May 2021

· A route into AMD engineering teams: technically owning and resolving issues for our partners and customers. Use of low-level engineering toolsets for greater understanding of performance bottlenecks. Submitting JIRA tickets and driving resolution.

· Automate repeatable procedures

· Document and publish performance results and procedures you have generated


Job Requirements

This position requires an individual with excellent verbal and written communication skills. The candidate for this role must be able to clearly demonstrate their application expertise in GPUs for HPC and it is anticipated the individual will have a minimum of 3 years HPC application experience in a recent role. This role requires the ability to independently manage multiple priorities to deliver results on time, and to build excellent working relationships both within AMD, across our network of partners, and our customers.

· Minimum 3 years CUDA and/or OpenACC experience developing HPC codes on GPUs, plus accompanying experience profiling and inspecting code for greater optimisation

· Experience with OpenMP

· Expertise in compiling large codes from source, with appropriately linked math libraries and flag optimisation

· Solid understanding of baseline testing of synthetic codes: HPL, STREAM, DGEMM, HPCG, HPCC

· Impact of system level hardware and its configuration/impact on performance

· Customer-facing experience. Able to write technical documents and communicate at an appropriate level depending on the audience (e.g. developers or commercial managers)

· Some Linux administration; understanding setup for HPC middleware.

· Bachelor’s degree in a technical discipline (computer science, electrical engineering, physics, mathematics)

· National and international travel. Initially about 10% travel but likely to grow to about 25% once this initial project is completed.

· Fluent in English


Like to have

· Over 5 years HPC application experience

· Experience working on very large codes such as weather/climate and associated tuning for greater scalability

· Any experience understanding/inspecting/writing assembly

· Understanding of memory and cache hierarchy and methods to query performance/latency at each level

· Understanding HPC dataflow down to the register-level

· Can speak German


Additional Expectations

· Role is based in the European time zone with periodic travel to the customer until summer 2021. There will be requirements to travel to other HPC Centers, customers, partners and relevant conferences both within North America and globally

· Some travel to India to work with our compiler and library teams




Requisition Number: 77405 
Country: Germany State: Bavaria City: Munich 
Job Function: Field Application Engineering/ Field Sup

Apply now

Apply for Job

Share this Job