UPMEM accelerates 20x big data applications with its Processing In-Memory chips, solving the dominant cost of data movement and the Memory wall, at marginal cost and energy consumption. UPMEM solution readily integrates in app servers to offer the 1st efficient scalable programmable acceleration solution for data-intensive apps such as genomics, AI, anaytics, search…
What you will do :
You will demonstrate the potential of Processing In-Memory on AI workloads.
You will adapt AI algorithms on the UPMEM Processing In-Memory PIM processing platform to showcase the acceleration potential of HW-based PIM solution, that is about to hit the market.
You will anlayze algorithms submited by UPMEM AI partners, work on the parallelization and porting of tke workload.
And participate to the PIM disruption.
What we are looking for :
Passionate about processing issues for critical apps, including AI, you are experienced in algorithm coding.
You have a good understanding of applications architecture, and you are aware of performance issues.
You will code in C and use the UPMEM SDK.
You will collaborate on a continuous basis with the R&D team in Grenoble and partners worldwide.
Skills and experience :
MS in software engineering or equivalent
5+ years in one of the following domains: Big-data algorithms and databases, Parallel programming and systems
Familiar with Agile methodology
Great oral and written communication and presentation skills in English
Fast, autonomous and rigorous developer, you enjoy facing and solving complex problems.