Teaching

Artificial Intelligence (AI) has already proven its potential as a new technological revolution. Through AI, information has become more accessible and can be precisely tailored to the requirements of designers and engineers. Furthermore, state-of-the-art Large Language Models (LLMs) elevate this data accessibility to an unprecedented level, as logical reasoning and methodical approaches are now also part of the capabilities of these models. 


These developments significantly influence existing teaching models at universities, as AI can provide the knowledge that students learn over years with high efficiency. To embrace this trend, our group offer a series of courses that systematically introduce the fundamentals of integrated circuits and hardware for AI, as well as AI-assisted circuit design methodologies. The concept of these courses is shown in the figure on the right. With these courses, students will not only become familiar with state-of-the-art AI tools but also understand the underlying scientific evolution. In all courses, practical tutorials and laboratory exercises will be integrated to allow students to gain direct experience in applying the lecture content. 

ReKI

Timing of Digital Circuits (Lecture+Tutorial, Lecture-focused, SS): Fundamentals of digital integrated circuits, race conditions and hazards, flip-flop-based design, setup/hold time constraints, clock frequency, gate library characterization, Static Timing Analysis (STA), advanced Statistical Timing Analysis (SSTA), techniques for performance optimization such as pipelining and retiming; clock networks and cross-domain clocking; asynchronous chip design. Moodle Course

Digital Design and Architecture (Lecture+Lab, Lab-focused, SS): Principles of hardware design; digital circuit architectures; design strategies of cryptographic accelerators considering performance and area; validation on FPGA platforms. Moodle Course

Hardware Accelerated Edge AI (Lecture+Lab, Lab-focused, SS 2027): Resource-efficient AI deployment; co-design of algorithms and systems; CPU and hardware accelerator synergy; system implementation and testing; validation on FPGA.

Resource-efficient Deep Learning (Lecture+Lab, Lecture-focused, WS): Acceleration strategies for neural networks; reduction of computational effort; logic design; in-memory computing; hardware implementation.

AI for IC Design (Lecture+Lab, Lab-focused, WS): AI-assisted system specification; automated code generation and simulation for circuits; EDA-tool-based optimization; design space exploration; LLM-assisted chip design; FPGA validation.