9 Nov 2026 Distributed LLM Inference on Kubernetes with llm-d, Wasm, & TornadoVM
WebAssembly and Tornado VM enable efficient local inference, while LLMD optimizes distributed GPU usage and semantic routing for on-premises AI deployment.
More Podcasts by InfoQ episodes

Published 29 Jun 2026
Duration: 00:38:56
Challenges in AI adoption highlight trade-offs between cloud-centric collaboration and local-first computing, with debates over data ownership, CRDTs for decentralized sync, hybrid workflows, and integrating AI with user-centric, offline-capable tools.
In this episode, Heroku co-founder and Ink & Switch founder Adam Wiggins argues for a 'local-first' architecture that reconciles cloud-based collabora...
The podcast explores challenges and innovations in technology adoption, emphasizing the long-term impacts of design decisions in AI systems and collaborative workflows. Discussions highlight tensions between cloud-based and local-first computing models, such as the trade-offs between accessibility and data ownership in cloud collaboration tools like Google Docs versus the need for offline capabilities and user control in local-first solutions. The evolution of version control systems beyond software developmentaiming to make tools like Git accessible for non-technical users in productivity applicationsis another focus, alongside critiques of existing tools for their lack of robustness. Advances in decentralized data management, such as Conflict-Free Replicated Data Types (CRDTs), are presented as critical for enabling real-time synchronization and conflict resolution in distributed environments, with applications in tools like Linears ticket tracking system. The conversation also examines the growing importance of hybrid models that balance local processing with selective cloud reliance, driven by edge computing and the need for flexibility in remote and low-connectivity scenarios.
Key themes include the historical underinvestment in traditional productivity tools compared to consumer tech, with recent innovations like Notion and Figma signaling renewed interest in enhancing collaboration and creativity. The role of platforms like Heroku in shaping modern cloud computing and serverless practices is discussed, alongside challenges of over-reliance on centralized cloud infrastructure, particularly in regulated industries. The "Local First" movement is framed as a response to cloud dependency, prioritizing user agency and data control while acknowledging the risks of geopolitical instability and centralized service outages. The discussion extends to the technical and philosophical balance between innovation and practicality, such as the tension between over-engineered infrastructure and delivering efficient value, as well as the potential of open-source, smaller models for on-device processing. Emerging technologies like CRDTs and decentralized protocols are positioned as foundational to future systems, though their maturity and adoption hurdles are noted.
Broader implications include the need to harmonize technological advancements with societal and regulatory considerations. The podcast addresses the contrast between the U.S. model of entrepreneurial risk-taking and Europes focus on privacy and data rights, advocating for a hybrid approach that balances innovation with protection. It emphasizes the importance of cross-regional collaboration and shared problem-solving, while reflecting on the challenges of integrating AI and local-first principles into existing frameworks. The conversation also touches on the evolution of conferences and communities uniting diverse stakeholdersfrom academic researchers to developers and designersto address open challenges in data ownership, user-centric design, and the ethical integration of emerging technologies. Ultimately, the dialogue underscores a recurring trend toward hybrid solutions that adapt to evolving needs without compromising user control or system reliability.
What if you reimagined a traditional productivity tool (e.g., spreadsheets) with AI-powered real-time collaboration and version control?
What if you built a hybrid app using CRDTs for local-first sync and selectively offloaded collaboration to the cloud?
What if you simplified version control for non-developers by integrating "pull request"-style workflows into word processors or calendars?
9 Nov 2026 Distributed LLM Inference on Kubernetes with llm-d, Wasm, & TornadoVM
WebAssembly and Tornado VM enable efficient local inference, while LLMD optimizes distributed GPU usage and semantic routing for on-premises AI deployment.
2 Nov 2026 Educating an Architect: A Conversation with Luca Mezzalira
Architects must evolve from technical designers to social-technical enablers, using domain-driven design to balance AI automation with human oversight in core systems.
12 Oct 2026 What Does It Really Mean To Program With AI: A Conversation with Joy Ebertz
"Explores software architects' evolving role in balancing technical and business challenges, AI's impact on modular architectures, risks of automation, and the need for human oversight in AI-assisted development."
5 Oct 2026 Future Cybersecurity: Hardware Memory Safety, Automated Governance and Post-Quantum Cryptography
"Challenges and advancements in AI, security, and software development are discussed, emphasizing governance, open-source maintenance, hardware security, automation, and regulatory compliance like the EU's Cyber Resilience Act, alongside post-quantum cryptography and AI's dual role in security."
28 Sept 2026 The AI Revolution Fails Without Psychological Safety For Developers: A Conversation with Erin Doyle
"Software architects must adapt to AI-driven shifts in system design, balancing automation with human oversight, team alignment, and ethical considerations in modern development."