This recap covers the Open Community for Research track at Open Community Experience 2026 (OCX26), focusing on how open source research moves from prototypes to open source platforms across IoT, AI, and digital infrastructure.
The sessions consistently moved from concepts to implementation. Whether the topic was IoT virtualisation, AI agent orchestration, or smart city platforms, the emphasis stayed on how research outputs are built, deployed, and maintained as open source systems.
Key topics from the Research track
A few patterns stood out:
- Research is expected to deliver working solutions, not just models or papers
- Open source collaboration under a vendor neutral governance model
- Many projects sit between academia and industry, creating gaps in ownership and long-term maintenance
- Interoperability across platforms, standards, and domains remains a limiting factor
Day 1 at Open Community for Research: Highlights
VOStack open source software stack for the virtualisation of IoT devices with Dr Anastasios Zafeiropoulos
Eclipse VOStack introduces an open source architecture designed to virtualise IoT devices across edge and cloud environments.
The system is built around the concept of a Virtual Object, which abstracts physical devices into programmable components. The architecture separates physical convergence, edge or cloud orchestration, and backend logic, allowing distributed systems to scale while maintaining control over device behaviour.
Dr Zafeiropoulos said, “the VOStack is an open source software stack that supports interaction with both physical IoT devices and edge or cloud computing orchestration platforms.” You can see the recording on the OCX YouTube channel.
One key takeaway: Research platforms are being designed as modular systems that integrate physical devices with distributed computing layers.
MOSAICO: management, orchestration and supervision of AI-agent communities with Massimo Tisi
This session focused on improving reliability in AI systems by moving from single-agent models to coordinated multi-agent systems.
Instead of relying on one model, MOSAICO proposes structured collaboration between agents that validate and challenge each other’s outputs. This reflects how human teams work, but with scalable coordination and lower cost.
The platform introduces orchestration, governance, and benchmarking capabilities to make these systems measurable and controllable. You can see the recording on the OCX YouTube channel.
One key takeaway: Multi-agent orchestration is emerging as a practical approach to improve reliability in AI-driven systems.
Live traffic, zero drama: Open source twins for smarter cities with Robert Hilbrich and Michael Behrisch
Digital twins for urban traffic are implemented using open source tools combined with real-time data streams.
The system combines live sensor inputs with continuous simulation updates, creating a feedback loop between real-world conditions and predictive models. This allows traffic systems to adapt dynamically rather than rely on static data. You can see the recording on the OCX YouTube channel.
One key takeaway: Digital twins are evolving into continuously updated systems that support real-time operational decisions.
Day 2 at Open Community for Research: Highlights
Seeds of openness: how higher ed grows the future of open source with Inge Donkervoort, Tonco Tijdeman, and Jan Muller
Universities continue to act as the origin point for many open source projects, providing space for experimentation and early development.
The challenge lies in moving beyond prototypes. The discussion focused on how initiatives help projects transition into maintained, production-ready systems that can be adopted outside academic environments. You can see the recording on the OCX YouTube channel.
One key takeaway: Universities generate open source innovation, but structured support is needed to move projects into sustained use.
UniTime overview: from research to practice with Tomáš Müller
This session provided a concrete example of a research project that became a production system.
The UniTime project evolved from an academic scheduling project into a widely used platform handling complex optimisation problems in real environments. It demonstrates how research algorithms can scale when they solve operational constraints directly. You can see the recording on the OCX YouTube channel.
One key takeaway: Research systems that address real operational constraints can transition into long-term production tools.
Aveiro Living Lab: A communication, sensing, and computing platform for smart cities with Duarte Raposo
A city-scale platform in Aveiro integrates IoT devices, communication networks, and data systems using open standards.
The architecture connects heterogeneous devices and enables real-time data collection across mobility, environmental monitoring, and public safety use cases. Interoperability is central, allowing different systems to work together without tight coupling.
Duarte Raposo highlighted that “the system connects 44 access points distributed across the city, enabling interoperability between heterogeneous devices.” You can see the recording on the OCX YouTube channel.
One key takeaway: Living lab environments are being used to validate research systems under real-world conditions at city scale.
Day 3 at Open Community for Research: Highlights
Open source strategies for Europe’s sovereign digital infrastructure with Rolf Riemenschneider and panel contributors
The final discussions focused on how open source underpins European digital sovereignty and industrial infrastructure.
The conversation brought together perspectives from policy, research, and industry to examine how open source can reduce dependency on external providers while supporting large-scale systems. Coordination across stakeholders remains a central challenge, particularly when aligning long-term strategy with immediate implementation needs.
Topics included governance models, risk management at scale, and what success looks like for European open source initiatives over the next decade.
One key takeaway: Building sovereign digital infrastructure depends on aligning policy, industry, and open source ecosystems around shared platforms and long-term coordination.
What this means in practice
Across the three days, a consistent pattern emerged. Research is expected to deliver systems that can operate beyond academia.
That requires:
- Designing architectures that scale beyond prototypes
- Ensuring interoperability across platforms and standards
- Planning for long-term maintenance and ownership
- Connecting academic work with industry adoption pathways
The gap between research and production is still there, but it is narrowing through open source collaboration and governance models designed to be used, extended, and maintained.
If you missed one of the sessions from the Open Community for Research, you can now see them on our YouTube channel.
