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 31 Mar 2026
Duration: 00:51:41
The text highlights the importance of learning from real-world failures over synthetic chaos experiments to build resilient software architectures, emphasizing systemic resilience engineering, adaptive strategies, and continuous improvement in complex systems.
In this podcast Michael Stiefel spoke to Lorin Hochstein about how real-world failures provide insight into how software systems actually work. Our fi...
The discussion centers on software reliability engineering, emphasizing how architects and site reliability engineers (SREs) approach system resilience differently. SREs focus on managing real-world failures and incident response, while architects design systems to inherently withstand disruptions. Key challenges include bridging the gap between AI proof-of-concept environments and reliable production deployments, as well as replicating complex, simultaneous real-world failures in synthetic chaos experiments. Tools like Chaos Monkey exemplify proactive resilience testing by simulating instance failures, but their limitations lie in their inability to fully mirror the unpredictable nature of actual incidents. The conversation underscores the value of learning from real system failures through incident reviews and post-mortems, which provide deeper insights into architectural shortcomings and human factors, rather than relying solely on controlled chaos experiments. Despite these efforts, knowledge sharing remains hindered by information overload and a lack of engagement from architects in applying lessons to system design.
The discussion also explores the trade-offs between system reliability and complexity, drawing parallels to biological systems like the immune response, which can cause unintended harm when overactive. Resilience engineering focuses on preparing for unanticipated failures through adaptive strategies, such as robust fallback mechanisms and dynamic scaling, rather than eliminating risks entirely. Feedback mechanisms in control systems, while critical for stability, can lead to cascading failures if mismanaged (e.g., resource saturation). Human and organizational challenges, including communication barriers and the limitations of individual accountability in large systems, further complicate reliability efforts. The conversation highlights the importance of systemic analysis over blaming individuals, advocating for cultural shifts toward blameless post-mortems and continuous improvement. Additionally, the role of complexity in software reliability is examined, noting that incremental changes and historical debt can undermine resilience, requiring ongoing adaptation to evolving environments and technologies. Finally, the dialogue addresses the inherent difficulty of software estimation and the need for systemic solutions to manage uncertainty, contrasting with the more predictable structures of fields like civil engineering.
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."