EPFL Research Data Week

By Prof. Berend Smit, Head of the EPFL Laboratory of Molecular Simulation
We argue that establishing a modular open-science platform centered on open, machine-actionable chemical data—building on existing technologies like electronic lab notebooks—would transform chemistry by making currently fragmented, often non-digital laboratory data reusable, discoverable, and integrally embedded in the research process rather than an afterthought.

By Dr Lucy Poveda, SIB Scientific Relations Manager
Breakthroughs like AlphaFold or the rapid COVID-19 response were only possible thanks to decades of investment in curated data resources. Yet these critical infrastructures remain underfunded and overlooked. This talk will show why sustaining them is essential to realise the FAIR principles and unlock AI-driven discovery.

By Prof. Martin Jaggi, Head of EPFL Machine Learning and Optimization Laboratory
How to build a 8B and 70B open-data open-weights LLM, multilingual in >1000 languages.

By Dr Anuradha Welivita, EPFL Human-Computer Interaction Group
This talk explores how empathetic conversational agents can embody fairness, empathy, and cultural inclusivity. It will also highlight how the quality, accessibility, and representativeness of datasets influence equitable AI systems, and how FAIR data principles may support culturally sensitive and transparent models, while acknowledging challenges such as protecting sensitive data and addressing gaps in dataset diversity.

By Dr Luc Patiny (maximum 20 participants) 
Discover how a modern Electronic Lab Notebook (ELN) is revolutionizing the management and sharing of scientific data in chemistry. This demo highlights a seamless workflow that integrates data collection, analysis, and reuse, making your research more collaborative and productive. With an accessible interface and powerful tools, explore a world where data becomes easy to find, use, and valorize. 
Prerequisite: Bring a laptop with Google Chrome installed. 

By Dr Guillaume Anciaux (maximum 20 participants) 
Solidipes is an innovative platform designed to support scientists throughout the entire lifecycle of their research data: acquisition, curation, publication, and sharing. This intuitive web solution facilitates data visualization, guides researchers in writing essential documentation and defining metadata, while ensuring compliance with FAIR principles (Findable, Accessible, Interoperable, Reusable). In this workshop, discover how Solidipes simplifies data management, removes barriers to sharing, and boosts scientific collaboration to fully valorize your research. 
Prerequisite: Knowledge of the Python programming language

By Dr Rok Roškar (maximum 15 participants) 
Join us to discover Renku, the open-source platform that enables scientific collaboration by bringing data, code, and compute together in a single accessible environment. Simplify your workflows, collaborate without technical barriers, and unlock the full potential of your projects with a flexible and powerful solution designed for researchers and educators. See how Renku is transforming collaborative research!  
Prerequisite : Knowledge of the Python or R programming language   

By Dr Francesco Varrato et Christopher Tremblay (maximum 18 participants) 
Who said publishing research data takes too much time or is too complicated? This hands-on serious game will prove the opposite! In small teams (2-3 people), you will explore the challenge of publishing a dataset on Zenodo and get a taste of the peer-review process. A prize awaits the best team! But the real reward will be discovering how fast and accessible data publishing can be – and how it can boost the impact of your research. 
Prerequisite: A laptop 

By João Prado Vieira and Claire Stoffel (maximum 20 participants)
A practical workshop to master storage location management, samples inventories, barcoding, derivations of samples and associated data.
Prerequisite: A laptop