Software Engineer (Open Source - 4TU.ResearchData)

Are you an enthusiastic software developer who wants to dive deep into reproducibility and research data and software workflows for machine learning and AI? Then we would love to meet you!

Software Engineer (Open Source - 4TU.ResearchData)

Job description

4TU.ResearchData is looking for an experienced Software Engineer to help strengthen and extend our repository for data and software. The 4TU.ResearchData repository uses in-house free and open source software called Djehuty. As part of our commitment to open science and the FAIR principles, we continuously work to improve the interoperability and reproducibility of data and software. Specifically for this purpose, we are seeking a Software Engineer who will mainly contribute to two projects funded by Open Science NL (part of NWO):

Enabling FAIR for AI: Infrastructure for Transparency and Reproducibility of Research Software

This project aims to make research data and software easier to reuse, verify and build upon. At present, software shared alongside research outputs often lacks the context and metadata needed to fully reproduce results. To address this, the project will improve transparency, interoperability and long-term usability by enhancing Djehuty, the open source repository platform used by 4TU.ResearchData and Nikhef. This work will support researchers by providing clearer guidance and more robust tools for responsibly sharing, citing and maintaining software, especially software associated with ML/AI models. Key project outcomes will include criteria for a graded badge to assess the quality and reproducibility of data, software and models, a software reproducibility service, and improved deposit, curation and reproducibility workflows.

Towards a community-driven and sustainable Research Software Directory (RSD)

The RSD is an open source platform that highlights the impact of research software using a broad range of qualitative indicators. This project aims to expand the RSD with a peer-review mechanism for software, improve curation support for communities and organisations, and link to ready-to-run cloud instances and reproducibility packages (being developed by 4TU.ResearchData) to stimulate software reuse.

In this role, you will collaborate closely with project partners from several Dutch institutions and work as part of the 4TU.ResearchData team at the TU Delft Library, contributing directly to the further development of Djehuty. You will coordinate development with other software engineers at 4TU.ResearchData to ensure alignment. You will also contribute to the maintenance and operation of the 4TU.ResearchData repository.

Your tasks and responsibilities:

• Guide and contribute to collaborative national projects, focusing on functionalities and workflows supporting reproducibility of software and AI/ML models;
• Research, design and implement solutions within the current architecture, extending functionalities where needed;
• Communicate design and implementation choices to project partners and the open source community, including updating documentation;
• Collaborate and coordinate development with project partners and the 4TU.ResearchData team to ensure alignment;
• Contribute to the maintenance and operation of the 4TU.ResearchData repository;
• Support other colleagues as needed.

About 4TU.ResearchData

4TU.ResearchData is a research infrastructure, community and training provider whose mission is to enable researchers from science, engineering and design disciplines to make a global impact with research data and software. We provide a trusted digital repository which support Findable, Accessible, Interoperable and Reusable (FAIR) data and software by offering curation, sharing, long-term access and preservation. The repository has a broad community of users and supporters who advance FAIR data and software practices. 4TU.ResearchData is governed and led by a consortium of four technical universities in the 4TU.Federation (TU Delft, University of Twente, TU Eindhoven and Wageningen University & Research) and is hosted by TU Delft.

Job requirements

• Strong proficiency in Python and its standard libraries;
• Proficiency with GNU/Linux, HTML/CSS, Javascript, Git;
• Full-stack web development awareness;
• Solid understanding of software engineering principles, including design, testing and version control;
• Comfortable working with complex systems, data workflows or research-oriented environments;
• Ability to translate requirements into technical solutions;
• Experience working with community-driven open source projects;
• Awareness of the FAIR principles;
• Good communication and team-working skills.

Desirable:

• Experience with research software, AI/ML workflows and model deployment;
• Familiarity with metadata standards, reproducibility frameworks and research data infrastructures;
• Experience contributing to or maintaining community-driven open source projects.

G.Kuhn@tudelft.nl; d.f.bangert@tudelft.nl