Background

We want to ease the AI model deployment on edge devices via compiler technology. Most frameworks (e.g., micro tensorflow) only allow to target a few devices or yield a large overhead from the runtime. Therefore, a novel and innovative solution is required. Be part of our path towards a more sustainable future of AI and participate in developing an AI compiler for RISC-V and neuromorphic systems.

Tasks

• Develop LLVM and MLIR extensions to map AI workloads on edge devices
• Build a runtime for RISC-V cores to execute ML models
• Demonstrate real-world applications to solve today's AI challenges

Supervisor

Dr.-Ing. Jan Moritz Joseph

Requirements

Must have:

• Good C++ and Python knowledge
• Interest in computer architectures and compilers