Computer Science
Work description
The work plan addresses the needs in current Research and Development (R&D) projects in INESC TEC to build energy-efficient software prototypes for training deep learning models on large-scale infrastructures (in particular, HPC centers). The specific objectives are detailed in the objectives section.
Academic Qualifications
MSc in Informatics Engineering, Computer Science or alike.
Minimum profile required
a) Knowledge and experience in the development of tools for energy consumption management.b) Knowledge and experience in the use of advanced computing services.c) Knowledge and experience with tools for training deep learning models.d) Knowledge and experience with the following virtualization/orchestration tools: Ansible, Singularity.e) Knowledge and experience with Linux systems.f) Language skills (written and spoken) for Portuguese and English.
Preference factors
g) Knowledge and experience with the deep learning frameworks PyTorch, DeepSpeed, and TensorFlow. h) Knowledge and experience with the tools nvidia-smi and RAPL. i) Proven experience in the design and implementation of benchmarking systems for assessing the energy consumption of applications.
Application Period
Since 20 Nov 2025 to 04 Dec 2025
Centre
High-Assurance Software