On February 21, 2025, 11am CET, Guido Juckeland from Helmholtz-Zentrum Dresden-Rossendorf will continue the HiRSE Seminar with his talk about Introduction of a quality indicator for research software publications
Abstract:
The Helmholtz Association is adding a new indicator for research data and research software publications to its reporting. A working group with members from all Helmholtz centers has been working on defining this new indicator and the Helmholtz general assembly has approved their suggestion in its fall meeting of 2024. In this talk the indicator as well as the ideas behind it and the methods to collect the information are introduced. The indicator is based on a maturity model looking at various aspects of a research software publication, thus also providing value to the authors of the software and research software researchers as it makes multiple aspects of research software as a scientific publication itself visible.
Institutions
On March 7, 2025, 11am CET, Samantha Wittke from CSC – IT Center for Science will continue the HiRSE Seminar with her talk about CodeRefinery: A Community for Teaching, Learning, and Growing Together
Abstract:
This talk will introduce the CodeRefinery project, its workshops, and its community, and explore how you can engage, contribute, and benefit from our resources as a researcher or Research Software Engineer.
CodeRefinery is a community-driven, publicly funded project dedicated to improving research code by providing hands-on training and open learning materials for researchers across disciplines. Our workshops focus on practical tools and “good enough” research software engineering practices that are often missing from traditional academic education—such as version control, reproducible research, collaborative development, and efficient coding techniques fostering Open Science and FAIR software development.
In addition to workshops and learning materials, CodeRefinery fosters a supportive and inclusive community where Research Software Engineers and enthusiasts can share knowledge, exchange experiences, and find opportunities for collaboration. This community-driven approach has enabled the joint organization of specialized workshops, such as high-performance computing (HPC) kick-off and “Tools and Techniques for HPC”, addressing a wide range of computational research needs.
Institutions
On April 2nd, 2025, 2pm CEST, Mascha Schmidt, Victor Haguet and Fritz Niesel from the Applied Machine Learning Group based at the Jülich Supercomputing Centre will join us at the HiRSE Seminar Series with their talk about Bridging the Gap Between Private Data and AI: A Beginner’s Guide to Retrieval Augmented Generation (RAG).
Abstract:
AI is transforming administrative and scientific workflows by automating tedious cognitive tasks. However, one of our most valuable resources - private and sensitive documents - often remains inaccessible to AI solutions. Retrieval Augmented Generation (RAG) offers a way to bridge this gap. It enables AI models to efficiently process and search large private data sources while maintaining control over data access and security.
In this talk, we’ll guide you through the process of building a RAG system, from understanding its core components to integrating it into real-world applications. We’ll begin by introducing RAG and demonstrating how it enhances knowledge retrieval where standard LLMs fall short. Then, we’ll break down key concepts such as chunking, embedding, vector storage and retrieval. Finally, we’ll explore how to tailor RAG for advanced use cases and integrate it into data-driven applications.
Whether you’re a developer, researcher, or AI enthusiast, this session will provide a clear roadmap for using RAG effectively. Join us to learn when to use RAG, what to expect, and how to take your LLM applications to the next level.
Institutions
Universität Hamburg
Adeline Scharfenberg
Universität Hamburg
Adeline Scharfenberg
Universität Hamburg
Adeline Scharfenberg