Research Engineer — HPC Memory Systems (Flexible Hours)

Barcelona Supercomputing Center; Barcelona; Sur place; EUR 50 000 - 70 000; Plein temps

Avantages offerts par ce poste

Restaurant tickets

Private health insurance

Relocation support

Flexible working hours

Training plan

Résumé du poste

The Memory Systems group within the Computer Science Department at the Barcelona Supercomputing Center is offering a full-time Research Engineer position to advance memory systems for AI. You will work on profiling, simulations, and prototyping on high-end hardware, addressing the memory wall in ML workloads.

Collaborations with a major memory manufacturer and industrial partners will provide opportunities to publish and contribute to cutting-edge HPC technologies such as HBM, DDR, CXL, and

Qualifications

• Good knowledge of computer architecture.
• Experience with Unix/Linux environments.
• Experience with profiling tools: Hardware counters, PAPI interface.
• Programming in C/C++, Python.

Responsabilités

• Profiling and performance modeling of HPC applications on high-end servers with advanced memory systems (HBM, DDR4/5, CXL).
• Exploration of Processing in Memory for HPC and AI applications.
• Prototyping of industrial products and tools.
• Close interaction and reporting to industrial partners.
• Benchmarking and performance analysis of high-end hardware products and prototypes.

Connaissances

Computer Architecture Knowledge

Unix/Linux Experience

Profiling Tools (PAPI)

C/C++

Python

Formation

BSc in Computer Science

MSc in Computer Science

PhD in Computer Science

Outils

PAPI

Hardware Simulators

Description du poste

The Memory Systems group within the Computer Science Department at the Barcelona Supercomputing Center is offering a full-time Research Engineer position to advance memory systems for AI. You will work on profiling, simulations, and prototyping on high-end hardware, addressing the memory wall in ML workloads.

Collaborations with a major memory manufacturer and industrial partners will provide opportunities to publish and contribute to cutting-edge HPC technologies such as HBM, DDR, CXL, and