Anuj Pathania

Anuj Pathania serves as an Assistant Professor in the Parallel Computing Systems (PCS) group within the Informatics Institute at the University of Amsterdam's Faculty of Science. His research pioneers sustainable computing systems operating under severe power, thermal, and reliability constraints, with significant contributions to energy-efficient hardware design and embedded systems.

Education:

• PhD in Computer Science (2018), Karlsruhe Institute of Technology
• MSc in Computer Science (2012), National University of Singapore
• B.Tech in Computer Science (2009), Maharaja Agrasen Institute of Technology

Pathania's research centers on low-power design and sustainable systems for constrained environments, with particular expertise in thermal management of 3D-stacked architectures and energy-efficient machine learning inference. His work bridges electronic design automation with real-world reliability challenges, developing novel power budgeting techniques like T-TSP that incorporate transient temperature effects ignored by conventional methods. Current projects include EU-funded initiatives on energy labeling for digital services, addressing ecological impacts through technological, behavioral, and legal frameworks.

His publication trajectory reveals a strategic evolution toward zero-waste computing, with recent work (2023-2025) focusing on hardware-software co-design for edge AI, energy modeling across computing continua, and parameter-efficient neural adaptation. Key themes include thermal-aware scheduling for S-NUCA many-cores, cooperative processor utilization in heterogeneous systems, and sustainability metrics for digital services.

Scientific Recognition:

• Best Paper Award Nomination at IEEE Computer Society Annual Symposium on VLSI 2023 for 3D-TTP power budgeting technique

Pathania actively mentors 4 PhD students (Ehsan Aghapour, Saeedeh Baneshi, Sudam Wasala, Yixian Shen) and has successfully supervised 5 Master's theses (including Cum Laude defenses by Joris op ten Berg and Jurre Wolff). His research is supported by major grants including Energy Labels for Ecologically Sustainable Digital Services (2023-2024) and Towards Zero-Waste Computing (2021-2025), developing simulation frameworks like HotSniper and CoMeT for thermal analysis. The PCS group maintains strong industry collaborations with ARM and NVIDIA, particularly through tools like ARM-CO-UP for heterogeneous processor utilization.