Context And Mission

The High-Performance Embedded Systems (HPES) laboratory aims at enabling the adoption of hardware, software, and artificial intelligent (AI) high-performance solutions in embedded systems as its center of gravity, but also in any system with some form of criticality, like cars, airplanes, and satellites. Our work is mainly carried out in the context of bilateral projects with several processor companies as well as several European-funded projects. For a complete list of publications of the group in the last years, please visit HPES page at the following link:
Modern Edge AI platforms integrate heterogeneous computing resources such as multicore CPUs, GPUs, NPUs, DSPs, and dedicated AI accelerators. Efficiently exploiting these resources requires mechanisms that go beyond standard application-level APIs and expose lower-level capabilities for resource allocation, partitioning, scheduling, performance monitoring, interference control, and runtime configuration.
In this context, we are looking for a talented and motivated Research Engineer to work on the advanced configuration, control, and performance optimization of hardware accelerators in cutting-edge Edge AI platforms.
The research activity will investigate how accelerator resources can be observed and controlled from the system-software layer, with particular attention to:
- runtime configuration and resource management of GPUs, NPUs, and other AI accelerators;
- hardware-supportedpartitioning and isolation mechanisms;
- execution-time andresource predictability for AI workloads;
- memory, cache, andinterconnect bandwidth management;
- accelerator scheduling and workload placement;
- power and performance configuration;
- hardware performance monitoring and telemetry;
- integration of low-level accelerator control mechanisms into system software, middleware, and runtime frameworks;
- experimental characterization of representative Edge AI workloads on state-of-the-art embedded platforms.
The work will involve both experimental platform development and research-oriented prototyping,



with the objective of identifying and exposing hardware and system-software mechanisms that enable predictable and controllable execution of AI workloads at the edge.
Candidates are required to exhibit a background in embedded systems and familiarity with both hardware and software analysis and configuration. Applicants are expected to hold an engineering degree in fields related to embedded systems, computer architecture, or closely related areas.
The position offers opportunities to collaborate with research institutions and industrial stakeholders across several European locations, providing a stimulating environment for professional growth, knowledge exchange, and hands-on research experience.

Key Duties

- Contribute to the research on advanced configuration and control mechanisms for GPUs, NPUs, and other hardware accelerators available in state-of-the-art Edge AI platforms
- Analyze hardware and system-software interfaces for resource allocation, partitioning, scheduling, monitoring, and performance management
- Contribute to system-level, application level, and containerization level solutions to improve non-functional metrics like execution time and resilience to hardware faults
- Develop low-level prototype software components, tools, and APIs to expose accelerator configuration and observability capabilities
- Document technical results and contribute to scientific publications, project deliverables, demonstrations, and presentations
- Collaborate with a small team of researchers and industrial partners working on embedded systems, AI acceleration, automotive computing, and software-defined platforms

Requirements

- Education
- Engineering degree or Master Degree in Computer Science, Computer Architecture, or a related field





- Essential Knowledge and Professional Experience

- Previous experience running AI workloads on embedded targets with accelerators like NPU and GPU

- Performance, heat, power consumption analysis of benchmarks running on embedded workloads
- Good programming skills in C/C++
- Solid knowledge of Linux systems programming
- Familiarity with computer architecture and heterogeneous computing platforms
- Experience working with embedded or edge-computing platforms
- Ability to perform experimental performance analysis and benchmarking
- Familiarity with scripting and automation tools, preferably Python and shell scripting

- Additional Knowledge and Professional Experience

- Embedded AI platforms such as NVIDIA Jetson/Orin, NXP iMX8, Qualcomm, AMD, Intel, or ARM-based SoCs
- Familiarity with Neural Processing Units and dedicated AI accelerators
- Linux kernel interfaces, device drivers, ioctl, MMIO, or low-level hardware interfaces
- Experience in programming and collecting hardware performance counters and accelerator telemetry

- Competences

- Problem-solving, pro-active, result-oriented work attitude
- Good communication skills including a good command of the English language (written and spoken)
- Good technical English and ability to work with technical documentation and research literature

Conditions

- The position will be located at BSC within the Computer Sciences Department
- We offer a full-time contract (35h/week), a good working environment, a highly stimulating environment with state-of-the-art infrastructure, versátil working hours, extensive training plan, restaurant tickets, private health insurance, support to the relocation procedures
- Duration: Open-ended contract due to technical and scientific activities linked to the project and budget duration
- Holidays: 22 days of holidays + 6 personal days + 24th and 31st of December per our collective agreement
- Salary: we offer a competitive salary commensurate with the qualifications and experience of the candidate and according to the cost of living in Barcelona
- Starting date: November

📌 Research Engineer – Advanced Configuration Of Edge Ai Accelerators (Barcelona)
🏢 Barcelona Supercomputing Center
📍 Barcelona