Workshop Organizing Committee:
- Lois Curfman McInnes, Argonne National Laboratory (ANL), chair
- Dorian Arnold, Emory University
- Prasanna Balaprakash, PrimaLabs
- Mike Bernhardt, Team Libra
- Franck Cappello, ANL
- Beth Cerny, ANL
- Anshu Dubey, ANL
- Denice Ward Hood, University of Illinois Urbana-Champaign
- Mary Ann Leung, Sustainable Horizons Institute
- Olivia Newton, University of Montana
- Keita Teranishi, Oak Ridge National Laboratory
- Stefan Wild, Lawrence Berkeley National Laboratory
Workshop Charge:
The high-performance computing (HPC) community has long driven scientific discovery at the limits of scale, complexity, and performance. Today, this leadership role is evolving rapidly as AI-enabled methods, heterogeneous architectures, and data-intensive workflows reshape how scientific computing is conducted. At the center of this transformation lies high-quality scientific software—the durable embodiment of domain expertise, computational methods, and collaborative practice that enables discovery to scale beyond individuals and institutions.
Motivated by urgent findings from recent community reports on scientific software development and AI for science, energy, and security, this workshop will convene cross-disciplinary experts spanning HPC, AI, computational science, applied mathematics, computer science, research software engineering, cognitive and social sciences, and community development. Participants will share, develop, and evaluate emerging strategies that address the challenges and opportunities shaping next-generation ecosystems for scientific computing.
The workshop will employ a co-design methodology, intentionally weaving together:
- Team-based scientific software development
- AI-enabled scientific workflows and software infrastructure
- Community, workforce, and ecosystem development
This integrated approach ensures that technical innovation and community evolution advance together, enabling broad, sustainable, and impactful scientific computing ecosystems.
This workshop represents Year 2 of a three-year series focused on strengthening team-based scientific software in an AI-driven future:
- Year 1 (2025): Identifying and understanding challenges, gaps, and opportunities
- Year 2 (2026): Sharing, developing, and evaluating strategies and progress
- Year 3 (2027): Coordinating ecosystem-wide advances that meld technical solutions and community-building
The 2026 workshop builds directly on insights from the 2025 workshop, as summarized in the report
https://doi.org/10.48550/arXiv.2510.03413, with an emphasis on leveraging synergies across AI, scientific software, and community initiatives to accelerate ecosystem-level impact.
Areas of emphasis: Our goal is to assess, strengthen, and transform scientific software ecosystems, informed by state-of-the-art team science, to meet the needs of next-generation research in scientific computing, while responsibly advancing emerging AI technologies. Technical and community perspectives will be integrated throughout discussions on the following themes:
- Software and next-generation science
- Examining how new scientific frontiers, AI-augmented workflows, and heterogeneous computing platforms demand new approaches to software and workforce development.
- Embedding community-driven practices to ensure that technical solutions reflect broad expertise, use cases, and stakeholder needs.
- AI-driven software ecosystems for scientific computing
- Mapping pathways toward robust, interoperable, and scalable software ecosystems that enable AI-driven discovery in HPC and data-intensive environments.
- Advancing AI tools, frameworks, and infrastructure for scientific computing that address current gaps while supporting openness, reproducibility, and broad participation.
- Team-based software and cross-disciplinary research
- Identifying evolving roles, career paths, and best practices for collaborative scientific software teams operating at the intersection of AI and computational science.
- Emphasizing community co-design, where software evolves through continuous dialogue among scientists, developers, users, and maintainers.
- AI for scientific software productivity and sustainability
- Exploring how AI-assisted development, testing, performance tuning, documentation, and maintenance can enhance productivity and long-term software sustainability.
- Embedding feedback loops that align AI-enabled tools with real-world scientific computing workflows and community needs.
- Community and workforce development
- Accelerating strategies to cultivate next-generation R&D teams equipped to operate in AI-rich scientific computing environments.
- Fostering collaborative cultures that bridge technical excellence with broad, community-centered ecosystem growth.
Workshop Objectives:
The workshop aims to shape a forward-looking, actionable vision for the future of team-based software in scientific computing, grounded in both emerging AI capabilities and the realities of cross-disciplinary collaboration. Key objectives include:
- Advancing scientific software practices: Building shared understanding of emerging methodologies, tools, and norms for team-based scientific software development, integrating technical rigor with community sustainability.
- Toward strategies for excellence: Curating and refining resources, frameworks, and exemplars that support excellence, effectiveness, and resilience in scientific software collaborations.
- Overcoming challenges and creating opportunities: Identifying strategies to address barriers faced by scientific software teams while enabling innovation aligned with emerging research and AI-driven needs.
- Envisioning the future: Articulating a shared, forward-looking vision for next-generation scientific software ecosystems that harmonizes technical innovation with community dynamics.
- Fostering community development: Defining concrete actions to strengthen and broaden the workforce, while fostering durable communities prepared to meet urgent challenges in HPC and AI-enabled scientific computing.
By intentionally integrating technical innovation with community-centered design, this workshop seeks to shape a future in which thriving, cross-disciplinary ecosystems drive the next wave of scientific discovery through advanced computing. In this future, high-quality scientific software—co-designed, AI-enabled, and sustainably maintained—serves as the keystone of enduring collaboration and scientific progress.
Workshop Outcomes:
Participants will share and examine perspectives, experiences, and emerging practices related to team-based scientific software in an AI-enabled future, with the goal of assessing progress, identifying remaining gaps, and prioritizing areas for continued attention. Through discussion and co-design activities, the workshop will surface high-impact focus areas, spanning technical, organizational, and community dimensions.
Insights emerging from the workshop will be synthesized in a post-workshop report, extending the findings of the 2025 workshop. This report will contribute to a growing body of community knowledge intended to inform ongoing and future efforts in scientific software development, workforce advancement, and ecosystem coordination, and may help to shape evolving perspectives, policies, and practices across the scientific computing community.
By intentionally interweaving community considerations with technical discussions throughout the workshop series, this three-year effort aims to expand and strengthen the scientific software community, foster cross-disciplinary collaboration, and help accelerate next-generation scientific discovery in an increasingly AI-driven world.
NOTE: Registration is by invitation only