Highlights from the Main Track at OCX26

The main track at OCX26 focused on the practical challenges of building and operating open source systems at scale. Across three days, sessions addressed software trust, digital sovereignty, distributed architectures, and the integration of AI into enterprise environments.

The discussions converged on a shared concern: systems are becoming more complex, more interconnected, and more exposed to regulatory and operational constraints. The sessions focused on how to design, validate, and operate these systems with measurable guarantees.

 

Key topics

Several themes defined the main track across all three days:

  • Trust as an engineering problem: moving toward measurable provenance, transparency, and accountability across software supply chains
  • Digital sovereignty through architecture: reducing dependency on external platforms through open source ecosystems and governance
  • AI in production systems: integrating AI into enterprise platforms while meeting requirements for security, reliability, and control
  • Distributed validation and orchestration: enabling reproducibility and scalability across edge and cloud environments
  • Operational maturity in open source: automating governance, measuring project health, and maintaining long-term sustainability

These topics were explored further in Birds of a Feather meetups and workshops. These sessions allowed participants to discuss implementation details, compare approaches, and examine trade-offs that are not always visible in formal presentations.

 

Other tracks at OCX26

The main track sits alongside the collocated events, each focused on a specific domain within the Eclipse Foundation ecosystem. These tracks explored the same challenges from different technical angles, with a stronger focus on domain-specific implementation and tooling.

 

Day 1 at OCX26: Main track highlights

Rebuilding trust: From open source to open accountability with John Ellis

John Ellis addressed the fragility of modern software systems and the limitations of current approaches to trust. He stated that “the objective of this session is to expose the growing fragility of modern digital infrastructure and to reframe how industry, regulators, and insurers think about software trust.”

The Eclipse Trustable Software Framework was introduced as a structured approach based on measurable provenance and transparent construction. Ellis emphasised that “trustable software engineering principles [require] measurable provenance, transparent construction, and shared accountability across the supply chain.” You can watch the recording of this session on our OCX YouTube channel.

One key takeaway: trust requires verifiable evidence embedded into the software lifecycle, not post hoc validation.

 

Digital sovereignty with FOSS with Wolfgang Gehring

Wolfgang Gehring focused on the technical implications of digital sovereignty. The session examined how dependency on external platforms introduces risk and limits control over critical systems. Open source was presented as a way to retain control over architecture, data, and the evolution of systems. You can watch the recording of this session on our OCX YouTube channel.

One key takeaway: sovereignty depends on architectural control and ecosystem participation, not policy statements.

 

The intelligent monolith: Supercharging Jakarta EE with local AI with Luqman Saeed

Luqman Saeed demonstrated how AI can be integrated directly into enterprise Java applications. The session showed that “this session demonstrates embedding AI directly within Jakarta EE applications, no external calls required.”

Using local models, the implementation achieved “zero-latency AI calls” and a complete RAG workflow within a standard Jakarta EE stack. You can watch the recording of this session on our OCX YouTube channel.

One key takeaway: AI capabilities can be embedded into existing enterprise systems without introducing external dependencies.

 

Day 2 at OCX26: Main track highlights

Edge-cloud architecture for scalable multi-simulator robotics validation with Massimiliano Vanini and Leonardo Montella

This session presented a distributed validation platform for robotics systems. It introduced “structured scenarios with measurable acceptance criteria” executed across multiple simulators.

The use of “cloud-native orchestration across edge and cloud resources” enables parallel execution and consistent validation across environments. You can watch the recording of this session on our OCX YouTube channel.

One key takeaway: distributed validation infrastructure is required to test complex systems at scale.

 

Beyond the mobile duopoly: open ecosystem, open language with Dan Ghica and Jaroslaw Marek

The session examined the concentration of control in mobile ecosystems. Eclipse Oniro and the Cangjie language were presented as alternatives designed for openness and interoperability. The session highlighted the role of programming languages in defining the constraints for AI-generated code. You can watch the recording of this session on our OCX YouTube channel.

One key takeaway: openness at both platform and language level is required to avoid long-term ecosystem lock-in.

 

Production-ready agentic AI: Building enterprise-grade Java systems with Jakarta EE and MicroProfile with Kenji Kazumura

Kenji Kazumura addressed the gap between AI prototypes and production systems. He noted that “the challenge often lies not in the AI technology itself, but in meeting stringent enterprise requirements for security, reliability, and transactional integrity.”

The session outlined patterns for building reliable AI systems using Jakarta EE and MicroProfile, focusing on secure integration and fault tolerance. You can watch the recording of this session on our OCX YouTube channel.

One key takeaway: production AI systems require the same engineering discipline as any other enterprise system.

 

Day 3 at OCX26: Main track highlights

How configuration-as-code eliminates invisible work for open source maintainers with Lukas Pühringer

Lukas Pühringer addressed the operational overhead in open source projects, describing it as “the ‘invisible’ work of managing repository security, governance, and compliance.”

The session demonstrated how configuration-as-code can automate these processes. “By shifting governance from manual enforcement to automated CaC… [organisations] eliminate countless hours of administrative burden.” You can watch the recording of this session on our OCX YouTube channel.

One key takeaway: automation of governance reduces operational overhead and improves consistency.

 

Level up your project: The open source challenge at the Eclipse Foundation with Maria Teresa Delgado

This session focused on sustainability in open source projects. She stated that “it’s not just the code, it’s the people, the collaboration, and the small habits that keep a project active and welcoming.”

The introduction of Project Score Metrics provided a way to measure project health based on activity, engagement, and transparency. You can watch the recording of this session on our OCX YouTube channel.

One key takeaway: project sustainability depends on measurable activity and consistent collaboration practices.

 

How Bloomberg uses Eclipse Temurin to power our enterprise Java software with Daniel Scanteianu

This session shows how Bloomberg deploys open source infrastructure across its systems, stating that “we use a variety of open source infrastructure, including Apache Kafka, Apache Solr, Apache Cassandra, and Apache ZooKeeper.”

The session showed how Eclipse Temurin is used across environments to ensure consistent and reliable Java deployments. You can watch the recording of this session on our OCX YouTube channel.

One key takeaway: enterprise adoption depends on standardised, reliable, and well-supported open source components.

 

What this means in practice

Across three days, the main track at OCX26 focused on how open source systems are designed, validated, and operated under real-world constraints. 

Several implications are consistent across the sessions:

  • Trust must be based on verifiable evidence embedded into the software lifecycle, including provenance, security practices, and transparency
  • Architectural control is required to reduce dependency risks and support long-term digital sovereignty
  • AI systems must be integrated into existing platforms with the same guarantees for reliability, security, and transactional integrity
  • Distributed architectures are necessary to validate and operate complex systems at scale
  • Governance, compliance, and project health must be automated and measurable to remain sustainable over time

These themes reflect a shift toward treating open source systems as critical infrastructure, where engineering discipline, operational consistency, and transparency are required at every stage.

If you missed one of the sessions from the main track, you can now see them on our YouTube channel.

 

Tags