The architecture, engineering, and construction (AEC) industry stands at a technological crossroads. While digital tools have transformed design capabilities, fundamental challenges in data sharing and collaboration persist across project lifecycles. Traditional file-based workflows create silos that hinder coordination, limit transparency, and constrain innovation. In a recent discussion on infrared.city Community Talks, Angelos Chronis invited Dimitrie Stefanescu, CEO and co-founder of Speckle, to explore how open-source data interoperability platforms are reshaping industry collaboration.

The conversation examines the socio-technical foundations of effective data sharing in AEC, revealing how community-driven development, transparent infrastructure, and federated data models are addressing longstanding industry challenges. Stefanescu shared Speckle’s evolution from the roots in recurring frustration to industry-wide platform, emphasizing that meaningful change requires both technical innovation and cultural transformation.

As the construction industry faces increasing pressure for digital transformation and sustainable practices, the insights from this discussion illuminate pathways toward more collaborative, transparent, and effective project delivery. The emergence of platforms like Speckle signals a fundamental shift from proprietary data silos toward open, interoperable infrastructure that serves the entire ecosystem.

Key points

• Interoperability demands socio-technical solutions that address both data exchange protocols and collaborative workflows. Traditional file-based systems fail to capture the iterative, co-creation nature of design processes, requiring platforms that support dynamic interaction and version control.

• Open-source development enables trust and extensibility through transparent code and community governance. This approach accelerates innovation while providing users with data sovereignty and customization capabilities.

• Federated object-based data models offer superior performance and flexibility compared to static file formats. Speckle’s approach enables real-time collaboration, granular change tracking, and semantic data relationships across diverse software platforms.

• Community-driven innovation proves essential for sustainable platform development. Academic-industry partnerships, contributor networks, and user feedback loops create resilient ecosystems that serve collective interests over proprietary control.

• Structured AEC data unlocks AI potential by providing machine learning applications with clean, accessible datasets. Unlike file-based approaches, object-oriented data enables sophisticated analytics without exploitative data capture practices.

The future of AEC collaboration depends on platforms that serve as neutral infrastructure rather than proprietary gatekeepers. Success requires balancing technical innovation with community stewardship, commercial viability with public benefit, and industry-specific needs with broader interoperability standards.

Continued below is an expanded analysis of the technical, social, and business dimensions of data interoperability in AEC, building on the key insights to offer practical guidance for implementing collaborative infrastructure in contemporary practice.

From Files to Streams: Rethinking Data Architecture

The Problem with File-Based Workflows

Traditional AEC collaboration relies heavily on exchanging static files—PDFs, CAD drawings, and BIM models—that represent snapshots of design intent at specific moments. This approach creates several fundamental limitations:

Information Loss and Translation Errors

• Each file export/import cycle introduces potential data degradation
• Complex geometric relationships become simplified or broken
• Semantic information (design intent, relationships) gets lost in translation

Version Control Challenges

• Multiple file versions create confusion about authoritative data
• Merging changes from different contributors requires manual coordination
• Change tracking remains limited to file-level timestamps

Limited Collaboration Capabilities

• Simultaneous editing requires complex file-locking mechanisms
• Real-time collaboration becomes technically infeasible
• Cross-disciplinary coordination relies on periodic file exchanges

Speckle’s Federated Approach

Speckle addresses these challenges through a fundamentally different data architecture based on object streams and federated databases:

Technical Advantages

Research from Stefanescu’s doctoral work demonstrates measurable improvements:

• 5x reduction in data exchange file sizes compared to IFC
• 2.47 average sub-classifications per model (vs. rigid IFC hierarchies)
• 18% user-defined object types (enabling domain-specific customization)
• 2.26 average receivers per data source (supporting collaborative workflows)

Practical Implementation

The object-based approach enables sophisticated workflows previously impossible with file-based systems:

• Granular Change Tracking: Individual geometric elements can be modified, tracked, and merged without affecting the entire model
• Semantic Preservation: Design relationships and parametric logic remain intact across software platforms
• Selective Synchronization: Contributors can subscribe to specific object types or project areas relevant to their work

Open Source as Competitive Advantage

Strategic Benefits of Transparency

Speckle’s success demonstrates how open-source development can provide competitive advantages in enterprise markets traditionally dominated by proprietary solutions:

Trust and Data Sovereignty

• Users maintain complete control over their data through self-hosting options
• Transparent code enables security auditing and compliance verification
• No vendor lock-in reduces long-term technology risk

Extensibility and Customization

• Community-developed connectors expand platform compatibility
• Custom workflows can be implemented without vendor dependencies
• Integration with existing tools becomes technically feasible

Accelerated Innovation

• Community feedback identifies real-world use cases and pain points
• Distributed development resources exceed single-vendor capabilities
• Academic partnerships provide research insights and validation

Community Governance Model

Speckle’s development illustrates effective open-source community management through its transparent governance structure, active contributor engagement, and balanced approach to commercial and community interests. This approach has enabled Speckle to build trust across different segments of the AEC industry, attract contributions from specialized domains, and establish resilient development patterns that support long-term evolution without compromising core values or user autonomy.

The platform demonstrates viable business models that support open-source development:

• Hosted Services: Cloud-based Speckle servers with enterprise support
• Professional Services: Implementation consulting and custom development
• Enterprise Features: Advanced analytics, security, and compliance tools

AI-Ready Infrastructure: Beyond File-Based Limitations

The Data Structure Challenge

Current AI applications in AEC often rely on image recognition or document parsing because structured design data remains trapped in proprietary file formats. This approach limits AI capabilities to surface-level analysis rather than deep understanding of design relationships and intent.

Traditional AI Limitations in AEC

• Image-based analysis cannot access semantic relationships
• Document parsing struggles with technical drawings and specifications
• File-based training data lacks temporal and collaborative context

AI Enablement

An object-oriented data structure provides AI applications with direct access to structured design information:

Rich Data Access

• Geometric objects include semantic properties and relationships
• Change history enables temporal analysis of design evolution
• Collaborative metadata reveals decision-making patterns

Ethical AI Development

An ethical framework for AI implementation in built environment data platforms would prioritize:

• Democratization of data sovereignty through user-controlled permission systems
• Methodological transparency in dataset curation and algorithmic validation procedures
• Distributed governance mechanisms that prevent centralized exploitation of collective knowledge

Building Sustainable Communities: Lessons from Speckle

Academic-Industry Integration

Speckle’s origins in the EU-funded InnoChain ITN network exemplify effective bridging between academic research and industry application, where doctoral projects directly address real-world industry challenges, while industry partnership provides crucial practical validation and feedback loops. This approach relies on commitment to open publication to ensure knowledge sharing extends beyond proprietary software development boundaries, creating a foundation for broad industry advancement.

Speckle’s independence from dominant industry players proved critical for long-term success; while early partnerships with major firms provided valuable support, complemented by multi-faceted community engagement across diverse platforms, including real-time communication channels, code collaboration repositories, user support forums, and knowledge-sharing events at academic conferences and workshops.

The Future of AEC Technology: From Data to Intelligence

The evolution toward AI-ready platforms demands fundamental shifts in how we approach technology architecture in the built environment:

-Federated data streams unlock distributed AI development. Structured AEC data from collaborative platforms provides clean training datasets for machine learning applications while preserving data ownership and privacy. Cross-project learning becomes possible when semantic structures align across the industry, enabling AI systems to identify patterns in design evolution that correlate with successful project delivery.

-Predictive analytics integration transforms project risk management. Time-series data from design collaboration enables predictive modeling of performance outcomes, cost implications, and construction risks. AI applications can analyze design changes in real-time to flag potential issues with energy performance, structural optimization, or code compliance before they become expensive problems.

-Platform ecosystems replace closed software solutions. Success increasingly depends on seamless data flow between design, analysis, and construction tools rather than feature completeness within individual applications. Technology distribution relies on integration quality and ecosystem compatibility, creating modular stacks that adapt to project-specific requirements and analytical workflows.

-Performance-based compliance accelerates regulatory evolution. Real-time monitoring capabilities enable continuous verification rather than periodic inspections, while building codes evolve toward outcome-based requirements supported by actual performance data. This creates opportunities for innovative solutions that meet environmental and safety objectives through novel approaches rather than prescriptive specifications.

-Real-time simulation integration becomes foundational infrastructure. Platforms like infrared.city demonstrate how fast environmental analysis during design phases transforms decision-making, enabling architects and engineers to iterate rapidly on performance outcomes rather than relying on post-design validation. This shift requires cloud-based computational resources and API-first architectures that connect design tools directly to simulation engines.

Conclusion and Strategic Recommendations

The conversation with Dimitrie Stefanescu reveals that effective data interoperability in AEC requires more than technical solutions—it demands fundamental changes in how the industry approaches collaboration, data ownership, and innovation development. Speckle’s success demonstrates that open-source platforms can provide viable alternatives to proprietary silos while creating new opportunities for community-driven innovation that extends beyond data sharing to transparent collaborative workflows that encompass performance analysis and environmental optimization at the core of design information model distribution. As digital transformation accelerates across the construction industry, the lessons from Speckle’s development offer crucial insights for building sustainable, collaborative infrastructure that connects seamlessly with analytical tools. The platform’s emphasis on transparency, community governance, and technical excellence provides a template for developing tools that serve collective interests, while enabling immediate feedback on design decisions through integrated simulation capabilities that make environmental performance as accessible as geometric modeling. The path forward requires careful balance between innovation and stability, commercial viability and public benefit, technical sophistication and practical usability. Organizations that can navigate these tensions while maintaining focus on value creation—and the ability to translate collaborative data into actionable insights about sustainability and performance—will be positioned to lead the development of truly collaborative AEC ecosystems that serve both coordination and optimization objectives.

Strategic Imperatives for Industry Stakeholders

For Technology Developers

• Prioritize interoperability and open standards over proprietary lock-in
• Invest in community development and governance structures
• Design business models that align with user interests and data sovereignty

For AEC Firms

• Experiment with collaborative platforms to develop internal capabilities
• Support innovative projects aligned with business objectives
• Build workflows that leverage real-time data analysis, simulation and collaboration and data sharing

For Educational Institutions

• Integrate collaborative technologies and data literacy into curricula.
• Develop essential environmentally lead design development literacy alongside with introducing real-time performance analysis methods
• Support research projects that bridge academic theory and industry practice
• Foster partnerships that enable knowledge translation and application between technology development and industry

For Policymakers

• Develop procurement frameworks that reward interoperability, transparency, and proven performance driven process
• Support technology development through research funding and pilot projects that demonstrate the value of connected collaboration and sustainable design platforms
• Create regulatory environments that encourage innovation while protecting user interests, particularly frameworks that enable real-time compliance verification through integrated design and analysis workflows

The future of AEC collaboration depends on building infrastructure that serves as neutral ground for industry-wide cooperation. Success requires moving beyond traditional vendor-client relationships toward community-based development models that create shared value and collective capabilities.

As Stefanescu emphasized, “We live to serve the industry, not hijack it.” This philosophy points toward a future where technology platforms act as facilitators rather than gatekeepers, enabling the kind of open collaboration essential for addressing complex challenges in the built environment.

Relevant Resources

• Dimitri’s PhD Thesis (UCL): https://dimitrie.org/thesis/
• Speckle Community Discussions: https://speckle.community
• Grasshopper Speckle Group: https://www.grasshopper3d.com/group/speckle
• Open-source repo: https://github.com/specklesystems
• Streamed interview: https://bricks-bytes.com/super-series-003-speckle/
• https://www.youtube.com/watch?v=jInlqA_mKjs
• https://bricks-bytes.com/super-series-003-speckle/