Manuel Eggimann

About

Manuel Eggimann is a leading researcher in energy-efficient artificial intelligence and Internet-of-Things (AI-IoT) systems, with a focus on ultra-low-power System-on-Chip (SoC) design. His most impactful work, a 2023 paper with 34 citations, introduces a groundbreaking AI-IoT SoC achieving 12.4 TOPS/W at 136 GOPS—a benchmark in efficiency. This chip integrates 16 RISC-V cores with precision-scalable deep neural network (DNN) acceleration (2-to-8 bit) and adaptive body biasing, boosting performance by 30% while operating within a few tens of milliwatts. Eggimann’s contributions extend to tactile sensing for electronic skin, where he designed an energy-efficient system for touch modality classification, reducing latency and power consumption in sensor-near processing. His work on the Marsellus SoC further advances heterogeneous RISC-V architectures for AI-IoT endpoints, enabling diverse tasks like augmented reality and personalized healthcare. With a total of 50 citations across his most cited papers, Eggimann’s innovations in precision-scalable DNN acceleration and adaptive power management are pivotal for next-generation, battery-constrained intelligent devices, positioning him as a key figure in the evolution of edge AI.

Research Focus

Key Achievements

Top Papers

1. 1
   22.1 A 12.4TOPS/W @ 136GOPS AI-IoT System-on-Chip with 16 RISC-V, 2-to-8b Precision-Scalable DNN Acceleration and 30%-Boost Adaptive Body Biasing

   34 citations · 2023
2. 2
   An Energy Efficient System for Touch Modality Classification in Electronic Skin Applications

   14 citations · 2019
3. 3
   Marsellus: A Heterogeneous RISC-V AI-IoT End-Node SoC with 2-to-8b DNN Acceleration and 30%-Boost Adaptive Body Biasing

   2 citations · 2023

Key Collaborators

Contact & Links